{"as_of":"2026-08-18T23:55:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:567e147a06c51e3656ae9d0217937f8183e4acce02e7c75f2f8ec194b4b6bc9e","coverage":[{"denominator":102,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":100,"source":"paper_references, paper_reference_links","source_observed_at":"2026-05-14T22:17:47.203589Z","state":"measured"},{"denominator":100,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":100,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-18T06:34:40.430872+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2603.27383/citation-record","integrity":"/paper/2603.27383/integrity","json":"/paper/2603.27383/citation-record.json","paper":"/paper/2603.27383"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"33f6dcc5-cf2c-4a03-98af-cafccb4344c8","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:671271ec7277c3ce5ae9e7d57f2f7757fccff7554709fbc801034d50fe6c82b1","observation_id":"f2bc25d3-01a2-4de8-a660-dc9706695a02","resolution":{"observed_at":"2026-05-14T22:18:05.055040Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Prune efficiently by soft prun- ing","venue":null,"work_id":"9dbcf353-f3f9-4b4f-8c11-570abbf57bc2","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:688cc6af0622dc67c60d0a57b1e2f8a1b0dc844485b54cd95193578fd5b612f5","observation_id":"f190d60e-3608-4f0e-9b23-e7fc1b190f83","resolution":{"observed_at":"2026-05-14T22:18:05.136724Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Deepcompress-vit: Rethinking model compression to enhance efficiency of vision transformers at the edge","venue":null,"work_id":"620eaa44-1ca5-42a5-b696-051bc000481b","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:e74b90ce8bc20ca81939de4105d9612223e12abd437b809f3959957187a29500","observation_id":"fca238d1-ba3d-4de2-8f97-f3a6a5745bb5","resolution":{"observed_at":"2026-05-14T22:18:04.942842Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Piqa: Reasoning about physical commonsense in nat- ural language","venue":null,"work_id":"4b9bd8fa-39a4-40bf-b3dc-af60b403fd83","year":2020},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:50fa0b938670ee8e0cdf6f008aabc8617da7361c338f60b97ee7faefb1540233","observation_id":"8aa1a071-a43d-44d4-83f6-f888ccf1d4cd","resolution":{"observed_at":"2026-05-14T22:18:05.141449Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Fcp dis vit: Efficient vision transformer with neural network pruning","venue":null,"work_id":"4813ed07-e9b6-4f78-afc8-7d25c172578c","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:c12acce25104df59eb884c341c720e35c4099e09631e8ffc070dada1b7b76afc","observation_id":"c4ae12fd-c781-44e5-a76d-a3859619f5f3","resolution":{"observed_at":"2026-05-14T22:18:05.040035Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"An image is worth 1/2 tokens after layer 2: Plug-and-play inference accelera- tion for large vision-language models","venue":null,"work_id":"61ca7bfb-2d07-43a3-9eaa-3fcdcec7031a","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:56974df9ad41c3d6759e9c85aa13ff49b2b1ca15a52df11427e8eda823edf8df","observation_id":"c0a415f4-c17d-4535-9e03-e5d836cb3168","resolution":{"observed_at":"2026-05-14T22:18:04.979344Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adaptformer: adapting vision transformers for scalable visual recogni- tion","venue":null,"work_id":"0c5d9c10-d3b4-47f9-bc8e-7df2b56223a0","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:fc6a24bfb5437c906fa930c24fa0b618febc8660b7bb27fc9d90f74301871050","observation_id":"694a06ce-13d2-4514-b753-c16b6d75a66e","resolution":{"observed_at":"2026-05-14T22:18:04.910505Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Lorashear: Efficient large language model structured pruning and knowledge recovery","venue":null,"work_id":"9ae3ceeb-cc65-4227-9ff9-2106fe1dc82e","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:1dd066cf5d19af162f113bbc049bd65afcf3d4c4f61ef491d4f8d16c44c5e84d","observation_id":"1ef58249-1e63-42fc-a574-166ba9733a3c","resolution":{"observed_at":"2026-05-14T22:18:05.073760Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"BoolQ: Exploring the surprising difficulty of natural yes/no questions","venue":null,"work_id":"e3998629-80ce-433d-a3ba-2aed8f043d90","year":2019},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:c4012da4a180f1c3147832a534f02e056e73c28aa067f9bd4feac06a5b9f0889","observation_id":"dce08176-ff00-4b4c-b621-fafa474762ad","resolution":{"observed_at":"2026-05-14T22:18:04.938306Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Think you have solved question answering? try arc, the ai2 reasoning challenge","venue":null,"work_id":"a38214e9-29ee-4f01-a2ca-2f5997a6b42e","year":2018},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:11ee3c9fdb00df0b19fd503fb922fb8af4a2e3c00fb8b4576c3d99516506c279","observation_id":"e27fd3f1-529c-40f5-baa5-45b32b35055c","resolution":{"observed_at":"2026-05-14T22:18:05.113598Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Imagenet: A large-scale hierarchical image database","venue":null,"work_id":"a3fe910c-f0bb-40cc-8d24-15f25816dd8a","year":2009},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:1db859cad191d87b282dba33b6d89a8fe02ee61cfe756a99cdb0307578c4b16f","observation_id":"045ea07c-c325-479b-9ef7-28c9644caef2","resolution":{"observed_at":"2026-05-14T22:18:04.888124Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Comedian: Self-supervised Table 10","venue":null,"work_id":"8f7f2567-4ed1-4ee5-a45f-0520c025b41c","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:46e52bfd5162fb3b1ae43263e151e673d70709f41e403916d27eb66a95026082","observation_id":"a968fecc-59bd-4946-90a8-8d8473ef9df7","resolution":{"observed_at":"2026-05-14T22:18:05.127063Z","resolver_source":"raw_fallback","status":"malformed_identifier"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2010.11929","last_updated":"2021-06-03T13:08:56Z","snapshot_observed_at":"2026-08-16T09:25:53.087782Z","submitted_at":"2020-10-22T17:55:59Z","title":"An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale","version":2},"cited_work":{"arxiv_id":"2010.11929","doi":"10.1175/jcli-d-22-0357.1","metadata_source":"pith","pith_arxiv_id":"2010.11929","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale","venue":"cs.CV","work_id":"e96730e3-129b-4db6-b981-15ab7932e297","year":2020},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"cited_paper":"/paper/2010.11929","citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:7dd9e1155f2ea95c5a497719789c35167e3af46d79c2ec42b7408492d3801bc4","observation_id":"9ec1fb50-bef5-435a-bc67-510021cfc5d7","resolution":{"observed_at":"2026-05-14T22:18:04.028168Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Sigmoid- weighted linear units for neural network function approx- imation in reinforcement learning.Neural Networks, 107: 3–11","venue":null,"work_id":"dcd649a2-c8dd-4265-842c-9d5c80113815","year":2018},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:c05bc5819bc1c3b85c98e5839919365e3c659f90555742ba94c898ce406bbe37","observation_id":"1ec9bf6c-df39-4a63-89e1-000db8793ef1","resolution":{"observed_at":"2026-05-14T22:18:05.223995Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Hyperdiffusion: Generating implicit neu- ral fields with weight-space diffusion.2023 IEEE/CVF In- ternational Conference on Computer Vision (ICCV), pages 14254–14264","venue":null,"work_id":"5e8faa39-58c2-4cb4-9c96-83fde4df7126","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:041ff24c529b4da1caaa8548bfaaaac7734adef39ada27febe0a5eef0f6fe797","observation_id":"08ecef17-8a3e-4573-b6d9-ff96379488b7","resolution":{"observed_at":"2026-05-14T22:18:05.245967Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Isomorphic pruning for vision models","venue":null,"work_id":"793a43ee-0557-45ee-b603-ca27c9979c22","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:227c919643dce71951cb192365c715774dd0c02046f87807fb42ad9f0aed0dfe","observation_id":"7aef1d85-e458-4e58-806b-413d0170699c","resolution":{"observed_at":"2026-05-14T22:18:05.237337Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"The language model evaluation har- ness","venue":null,"work_id":"466499bc-c3c4-4cd9-a6e3-46dcf9b378c8","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:23a66404e9c543cb83a1761d86d094f6554482b405e41f3fec881291de1b5b9e","observation_id":"9ce66074-28ab-4f40-b083-b3f7e8a76cbf","resolution":{"observed_at":"2026-05-14T22:18:05.205712Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Fast r-cnn","venue":null,"work_id":"56ca2411-9dd3-4dcd-9f3a-b77c6baebbad","year":2015},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:428672b8fb121dfd65710e9b0bd77a94e02d84b4a0f2d4870797ae40399c9c93","observation_id":"109f4da0-e952-45b6-83c3-1be4b85f6e29","resolution":{"observed_at":"2026-05-14T22:18:05.214870Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Pruning by block benefit: Exploring the properties of vision transformer blocks during domain adaptation","venue":null,"work_id":"38bf188d-a328-4990-9621-f8aa0438199e","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:db6db2b2d820cbf7ac2472ed4a4e75f70e448de12a0ae112ae46471e8deeda64","observation_id":"2ea0aaf7-5193-4f65-be64-e365eb8825f3","resolution":{"observed_at":"2026-05-14T22:18:05.220023Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Deep sparse rectifier neural networks","venue":null,"work_id":"fa41b7e5-ffe2-47ee-9411-6f66f8121e36","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:a26b996f13d23dcb6f0a3cdbe24dc5cb4ada4521f57c46b73b847e73ebd6ad3c","observation_id":"b08c0f08-772c-4d22-a59f-e6f40d6360ad","resolution":{"observed_at":"2026-05-14T22:18:05.210295Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Balf: Budgeted activation- aware low-rank factorization for fine-tuning-free model compression","venue":null,"work_id":"18a82e7c-b7eb-4ebf-bf43-d3dcf4417e16","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:dcc481ec9d5b9719bd72c79cdbaa09b0be2b83188218551a45b6073806bc7678","observation_id":"c2ab5e6b-d913-45f5-a92b-33e107ed5b86","resolution":{"observed_at":"2026-05-14T22:18:05.201265Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"The llama 3 herd of models","venue":null,"work_id":"e489b9ec-b919-4b6a-8ea4-edfed51d482a","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:d1e60c9fd4955baa4add475abd38c61655532e11d9277ed787f6f75d5e943617","observation_id":"f67f741d-5a1d-4885-8c17-aace5adc09fc","resolution":{"observed_at":"2026-05-14T22:18:05.241471Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Weight copy and low-rank adaptation for few-shot distillation of vision transformers","venue":null,"work_id":"7ae6dfc3-2cfd-4c2e-b533-a2852ebcd76b","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:907f11a48e36a63640250aa0f07a2638b3a59f3b5d6c2db0e5d67ed6f7131ba5","observation_id":"1b8b5719-8276-4e51-bf79-269160a0414d","resolution":{"observed_at":"2026-05-14T22:18:05.228055Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Dai, and Quoc V","venue":null,"work_id":"12db1923-4b17-4a75-a1bc-01c30e73f248","year":2017},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:019b3362731d597532b09748d367d1365f16dd57a60496cfc2e33a0bcf2d17a2","observation_id":"cd140a09-4bf3-44d2-8a69-48a66476f993","resolution":{"observed_at":"2026-05-14T22:18:05.187504Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Svdiff: Compact pa- rameter space for diffusion fine-tuning.2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","venue":null,"work_id":"6aba4240-2545-49e3-a6de-5cd321dec0eb","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:de26fe8faf8bfe46d54fdd65e5e7593886d0cf3b25cc9bd687287d3743634484","observation_id":"c84429ab-a114-4d92-b370-86c60c8b23f8","resolution":{"observed_at":"2026-05-14T22:18:05.183027Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Learning effi- cient vision transformers via fine-grained manifold distil- lation","venue":null,"work_id":"ab404ac0-481e-4700-9abb-ce5e296842f3","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:9ae8a53a673a18b3d9969fbaf6cb86ad96a4603067ed827bd42296072713cc50","observation_id":"af7291b4-73a5-4446-bca6-dbe1720ced32","resolution":{"observed_at":"2026-05-14T22:18:05.250653Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Multi-dimensional model compression of vision transformer","venue":null,"work_id":"a17e1872-4429-4180-9fb1-fad77d81fbcf","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:7adbd35836771be8bd083a0ad45364c4a625ac85c40a5abdc152e36211e94c04","observation_id":"cd6b4657-d6d9-488d-96c3-528d729e50cb","resolution":{"observed_at":"2026-05-14T22:18:05.191877Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"LoRA: Low-rank adaptation of large language mod- els","venue":null,"work_id":"65aa936a-baab-4644-b75d-f35817192a59","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:4f8add08ac045509a29404522ad321ecf8317b86934514c994a0ff37fe1360b6","observation_id":"737ffce7-078f-4cf3-818a-8e40129dd0cd","resolution":{"observed_at":"2026-05-14T22:18:05.196465Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"LLM-adapters: An adapter family for parameter-efficient fine-tuning of large language models","venue":null,"work_id":"ede01a8d-07dc-4182-a614-7fa4202a3bcf","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:5e786ae66b29960d9a0c06ec217fbfd48e3b1649ad614f1f15589b0ae47f9626","observation_id":"69f54d19-4135-432e-b8ec-de075e2e8f85","resolution":{"observed_at":"2026-05-14T22:18:04.974990Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"HiRA: Parameter-efficient hadamard high-rank adaptation for large language models","venue":null,"work_id":"3f547d33-8cea-4b3f-bfc0-40a15644afd0","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:ae4cfaae6be2e715a6f107baa625e5626d87228251b1c464bc6f01622caf7a68","observation_id":"9477980c-f17c-43d8-a5f0-68f59d620463","resolution":{"observed_at":"2026-05-14T22:18:04.803961Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Finding lottery tickets in vision models via data-driven spectral foresight pruning","venue":null,"work_id":"6d49aa52-0760-4387-84f2-77cec7c751e7","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:64b2276e4ff378789f4eb1a30bf68a45743e59055a6a555a69c1a759e30bde72","observation_id":"c9b88a66-92dc-4cc0-abef-009b99a65b0b","resolution":{"observed_at":"2026-05-14T22:18:05.132257Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adap- tive class token knowledge distillation for efficient vi- sion transformer.Knowledge-Based Systems, 304:112531","venue":null,"work_id":"f41afdb5-d575-4646-8826-868a5c4d25ef","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:a1cab0018dd48eae88f198bff2770582f7bbf638e659e3714ccf43aab8e0b549","observation_id":"a682978b-0e23-409c-87f1-a0e426e85ce8","resolution":{"observed_at":"2026-05-14T22:18:04.859504Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"The need for speed: Pruning transformers with one recipe","venue":null,"work_id":"932c2276-5029-4cff-ad66-749c09e4b936","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:75097661459ed0062a6820fd69f118b859b848ec23988b4114b0f621eff0e45e","observation_id":"e5728fed-01aa-4fa7-95c4-ebfb13ec8bd1","resolution":{"observed_at":"2026-05-14T22:18:04.960196Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Krizhevsky and G","venue":null,"work_id":"dcf3526b-1096-4e28-afab-3f239d8b7d63","year":2009},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:30cf031e0ac524f5e264e4861e83fa81acf7a00e75fba20ad346d2232019da6e","observation_id":"209f5365-c068-4e52-8d5d-fa9b78abebe9","resolution":{"observed_at":"2026-05-14T22:18:04.836407Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Dct-vit: High-frequency pruned vision transformer with discrete cosine transform","venue":null,"work_id":"65c33c70-a6e1-4d36-b195-d7dc213a0c62","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:627d57d2b12fbc8bf3976881cc097cc5eed205dcc5d7354d86b857789e48352d","observation_id":"eb5c34ae-c085-45e7-bec8-a1c279e90bfc","resolution":{"observed_at":"2026-05-14T22:18:04.947139Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Discovering sparsity allocation for layer- wise pruning of large language models","venue":null,"work_id":"eaf5dd55-3a84-445d-89f2-2744ead0faf3","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:b5ebea110dfb7806738b9a64c13187d31960f7bfd7c7487627134b5c676b36e9","observation_id":"a15e4eed-5b5c-4688-a804-da3f66e0600d","resolution":{"observed_at":"2026-05-14T22:18:05.077832Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Vb-lora: Extreme parameter efficient fine-tuning with vector banks","venue":null,"work_id":"dd14a399-3c93-49e9-84cd-9626305b9d0a","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:c9c6110d2ddbfe4a1b9f3c113eb7c89c906d40b8e2438db3056ce2d2cedc25e2","observation_id":"dbf86509-4509-4d68-a7ac-df81cd7cb434","resolution":{"observed_at":"2026-05-14T22:18:05.025947Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"LLaMA-VID: An image is worth 2 tokens in large language models","venue":null,"work_id":"ede6ad6b-085e-447a-8373-996898144456","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:d325032b62465cb7a419d7f542928e2ed2d1159501c98119618981ce35d93887","observation_id":"78155340-3b86-4200-aef5-d76ecda1994f","resolution":{"observed_at":"2026-05-14T22:18:05.016619Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Scaling & shifting your features: A new baseline for efficient model tuning","venue":null,"work_id":"8fe02859-90a8-43ab-b226-24bc8ecd8a16","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:90695191db76d75cdcffbcee6562e36d241b930776777ee7a6b306d01b9d3e29","observation_id":"cb0e29c2-e31c-43ac-9eb0-06bb3ee86597","resolution":{"observed_at":"2026-05-14T22:18:05.154887Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Inflora: Interference-free low-rank adaptation for continual learning","venue":null,"work_id":"ff33a10b-7350-4f85-acb7-18f76b2040de","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:9ce7709fc07d445e74fbd80007be06433933194cc7f81738d18a0e05cfedab49","observation_id":"fb591250-45f6-42c8-992f-f67c05bef274","resolution":{"observed_at":"2026-05-14T22:18:05.069058Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"3-in-1: 2d rotary adapta- tion for efficient finetuning, efficient batching and compos- ability","venue":null,"work_id":"d394343c-1132-4e2b-9db7-7ea0ba42dc77","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:78d288e85cd1f0c115b394a4b1e4408761fbb49e33ab72d6bd94999f5b83fd74","observation_id":"46bec15a-e612-431d-a25e-046e3af86522","resolution":{"observed_at":"2026-05-14T22:18:05.009842Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Slimgpt: Layer-wise structured pruning for large language models","venue":null,"work_id":"153d338e-10f1-4bf4-bd0d-5225602dc849","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:ea50abff9a0a1604dc8f996ed79b9f8a28a60f26e347cc10edc3637907c53598","observation_id":"8314b43e-193a-4184-b74a-c2cf6dd41610","resolution":{"observed_at":"2026-05-14T22:18:04.794762Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"SVFT: Parameter-efficient fine-tuning with singular vectors","venue":null,"work_id":"f4ad8d42-da28-484b-b706-b7c8867d72f5","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:31e04204c55bc1b08336673e273809fb6f0b8aa2d3c1426bf351b9dc65a69838","observation_id":"4053e72c-7995-423d-a698-a2b05d22ef8f","resolution":{"observed_at":"2026-05-14T22:18:04.831871Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Pruning via merging: Com- pressing LLMs via manifold alignment based layer merg- ing","venue":null,"work_id":"787de07b-8295-4e75-9b02-8652ab17651a","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:dcfe73f32ea960e7b9bbba19c9a05204a9ccd447a76893514ca692fafcd2542f","observation_id":"2ecb8b00-0049-4748-8c5a-05521c3e1f16","resolution":{"observed_at":"2026-05-14T22:18:05.087364Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning","venue":null,"work_id":"7ee88d4c-42d4-4d58-a580-4fd0d35cbfde","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:0d27b645fd06d627145aaf5d4bc6f3af7faae9953732305c9decbfd711ea0f07","observation_id":"a7d838b8-7ca2-4a04-b747-a55ddda86988","resolution":{"observed_at":"2026-05-14T22:18:05.082659Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Small scale data-free knowledge distillation","venue":null,"work_id":"3af38458-6e3f-460d-b699-790814abe0b3","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:96eb66dc9440931873e0b436d1ce79d5dc18be82fb8eba89b2aa33f5e4120c5c","observation_id":"4578e04c-df38-4407-b995-c4dd6450d2b1","resolution":{"observed_at":"2026-05-14T22:18:05.099622Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Dora: weight-decomposed low-rank adaptation","venue":null,"work_id":"f27369f1-1403-4c57-b580-35281f237e41","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:4723931e365873848d8ebf0734dbf721d9a9ef8b31de8dd9e4f27aed4a0d5cda","observation_id":"7a43ddb2-7670-41ee-84fa-34fc6f306dc5","resolution":{"observed_at":"2026-05-14T22:18:04.808805Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Black, Adrian Weller, and Bernhard Sch ¨olkopf","venue":null,"work_id":"2d7a12c6-2ca4-42b3-9551-58c76c239800","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:5ba6cb04c44e93c71013e3946591035834388e2da10cb2f081348cdb5f27d606","observation_id":"961bbaf0-ec23-45dc-9ae9-9a80bf1d8e29","resolution":{"observed_at":"2026-05-14T22:18:04.991992Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Decoupled weight de- cay regularization","venue":null,"work_id":"377cdef7-ac11-4004-825f-a768bbae580a","year":2019},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:6998c1a5f4d75f0bc12124751b3ab175255bc38391b3f1d150be4fb73ada2b46","observation_id":"af36c8cc-4fe4-491a-802b-7672cb0243d7","resolution":{"observed_at":"2026-05-14T22:18:04.933859Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"9d8d4e9b-dadc-4306-9aa7-b3f3be974d40","year":2013},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:28eb29989708d8156c841bb6394485afa71d3a9ef565d793606cafee487c3551","observation_id":"3ce3e647-972c-4caf-9ac8-8c170ce63a26","resolution":{"observed_at":"2026-05-14T22:18:04.854727Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Prune and merge: Efficient token compression for vision transformer with spatial in- formation preserved.IEEE Transactions on Multimedia, 27:4670–4683","venue":null,"work_id":"6dbc8736-b8d7-4608-ad30-ab97aee7f26b","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:5a432477dbb85292f7df8b3b025577e4aa51d79c5e1ec9c6ea01ef08ddecb469","observation_id":"40a00d95-5495-4d0a-a373-44afca1c5838","resolution":{"observed_at":"2026-05-14T22:18:05.178160Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Can a suit of armor conduct electricity? a new dataset for open book question answering","venue":null,"work_id":"0afec2d3-eda8-4823-81a2-4cabea3c97d5","year":2018},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:4071b2913d8dc2a92c720db17432b980b11ceb4decadf66f475e455c7339a818","observation_id":"b98c801d-73ad-4c73-a0fe-a2cfe2ab992d","resolution":{"observed_at":"2026-05-14T22:18:04.924330Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"RoSA: Accurate parameter-efficient fine-tuning via robust adaptation","venue":null,"work_id":"9ec94cb7-ddc3-4b68-9ec8-f1090299a1e9","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:8d334c3481c2f78a47fef87cd6ccdef89d59c476d3fd56c5881e1d2b36db416f","observation_id":"9191b26b-dd0c-46c8-927f-4e22ccaeee0c","resolution":{"observed_at":"2026-05-14T22:18:04.905533Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"RoSA: Accurate parameter-efficient fine-tuning via robust adaptation","venue":null,"work_id":"fdf77681-9ecf-47ff-be36-223245ec0278","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:db51269064b5b3df0a3314c15c41e8daf550db1643cc38867eba68f6b2799f41","observation_id":"73c1fa0a-49da-4d73-818f-73ae7f44ba6c","resolution":{"observed_at":"2026-05-14T22:18:04.928088Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Automated flower classification over a large number of classes","venue":null,"work_id":"6c705ee8-8942-4b65-b4e6-05266d53e7e8","year":2008},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:78415e275ff471df7ac167d157a5d7c1816faa55d630a62f01d9d16732e1ff1a","observation_id":"a5ea0b6b-2784-4edd-a99c-36d45f401cec","resolution":{"observed_at":"2026-05-14T22:18:05.021512Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Fair-vpt: Fair visual prompt tuning for image classification","venue":null,"work_id":"0df2e90b-349d-4e14-8af7-f2f2bdd25bca","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:0c98da5d8c52d7051311503af98fc99a008ee7898f2c22d44592e6fd7dea941a","observation_id":"2f34a2f6-39c0-47bf-98e7-f6c03154abbc","resolution":{"observed_at":"2026-05-14T22:18:04.874748Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.16319","last_updated":"2025-02-24T07:41:18Z","snapshot_observed_at":"2026-08-16T14:15:26.382683Z","submitted_at":"2024-02-26T05:51:47Z","title":"Data-free Weight Compress and Denoise for Large Language Models","version":2},"cited_work":{"arxiv_id":"2402.16319","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2402.16319","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Data-freeweight com- press and denoise for large language models.CoRR, abs/2402.16319","venue":null,"work_id":"a16e0f28-9711-41ba-8a21-8e1246bae48c","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"cited_paper":"/paper/2402.16319","citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:2cd5047a9e5379d9c6549402d840619fd39c4e581117acf50b41f8780cff03cf","observation_id":"7ec1d2a2-cfa1-4d9f-96bf-1027a2a35df6","resolution":{"observed_at":"2026-05-14T22:18:03.999999Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Plummer, Nikoli Dryden, Julius Frost, Torsten Hoefler, and Kate Saenko","venue":null,"work_id":"88cfd363-b88e-4bed-a7b0-521074ed61cd","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:9b2469392fb841e48dea1c4366d0b4ae19a3e911f011aea57fa77059951b1c20","observation_id":"b50dce33-ff2d-4524-83ec-a8016949c9bf","resolution":{"observed_at":"2026-05-14T22:18:04.850558Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Recognizing in- door scenes","venue":null,"work_id":"121737a8-f1e3-495a-9c3a-3931a5fa94b5","year":2009},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:565ddca49c449e27c495e4a7b0be2308b7c82dd8dde789e2c275fff1d5f459f5","observation_id":"22b8286a-bce4-4fde-a96f-2df8ca337ee2","resolution":{"observed_at":"2026-05-14T22:18:05.232754Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Venkatesh Babu","venue":null,"work_id":"89677cd6-57b3-45a7-aaf7-be43f5181ace","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:36b15b3770f30100700ca8015a95afe829b357fa6f953862207122d061a6d342","observation_id":"89790b05-372d-4626-bea6-ed32b9804af9","resolution":{"observed_at":"2026-05-14T22:18:04.840991Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"MELoRA: Mini-ensemble low-rank adapters for parameter-efficient fine-tuning","venue":null,"work_id":"2980c4b6-7ba9-47a8-a087-368d72b13cfd","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:0dfad6e2ad522c0ba93493e751ebd6e9805cb8ae69f12fc5164e55872bbe740b","observation_id":"a61fd0e5-3ccf-4f7b-901e-8e569e8ca8eb","resolution":{"observed_at":"2026-05-14T22:18:04.996245Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Winogrande: an adversarial winograd schema challenge at scale.Commun","venue":null,"work_id":"1e1a9174-6657-4164-8afd-351aca1470ae","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:e7d4f16d0c8bf5ac7f60d252ecde990f6835d03a8f9f257ae17fc99e61e0af70","observation_id":"e70e5054-4513-47d7-a255-320687bdebde","resolution":{"observed_at":"2026-05-14T22:18:05.004782Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Social IQa: Commonsense rea- soning about social interactions","venue":null,"work_id":"ab55c7ec-ac32-4814-95c5-a63cfb3a5470","year":2019},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:04c247760c57580310288ecaa910bfb2c8db848f4d799400146d850627f0738b","observation_id":"f92ea45a-e7e4-448e-80d4-f6eaee0220ab","resolution":{"observed_at":"2026-05-14T22:18:05.150292Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Learning implicitly re- current CNNs through parameter sharing","venue":null,"work_id":"b61ea12e-325f-4a28-9b4b-c36676e3eba9","year":2019},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:b175a8404554859d9e50d5db2efc5194561ba997d4afd740b1ebd4cfb02c6c16","observation_id":"3511ae88-7de0-42cb-92d6-4f30194bb6e3","resolution":{"observed_at":"2026-05-14T22:18:04.900654Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"You only prune once: Designing calibration- free model compression with policy learning","venue":null,"work_id":"42fcecba-d71e-43cf-94f5-d607f79db358","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:fccb1c8bf8434ee4b1ffcfd7290a17e5d6aef11c33e5a21046653512931da26c","observation_id":"b3a04812-57df-4dd4-aea2-da69ced4f808","resolution":{"observed_at":"2026-05-14T22:18:05.108993Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Pro- vpt: Distribution-adaptive visual prompt tuning via prompt relocation","venue":null,"work_id":"90356480-82e8-41c9-b412-ccb4e35208c8","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:c4409cdf9a0aa0dfff554c1e68a66a23a8959afce72f8c58c52daca013f252fc","observation_id":"446a5487-f2f6-423b-aac0-f694ef964664","resolution":{"observed_at":"2026-05-14T22:18:05.063820Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2506.07138","last_updated":"2025-06-08T13:36:06Z","snapshot_observed_at":"2026-08-07T05:39:00.952778Z","submitted_at":"2025-06-08T13:36:06Z","title":"Learning Compact Vision Tokens for Efficient Large Multimodal Models","version":1},"cited_work":{"arxiv_id":"2506.07138","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2506.07138","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Diversity-guided mlp reduc- tion for efficient large vision transformers","venue":null,"work_id":"5edd8a84-f66a-46b9-be7d-a5eefdbdeaf3","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"cited_paper":"/paper/2506.07138","citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:013745175a0f69350cf9bc83d95f35984c1efb6c49a5222510ed8bbc83720fb5","observation_id":"cb49eef2-ff18-401d-a49c-9f8578789311","resolution":{"observed_at":"2026-05-14T22:18:04.014300Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"UPop: Unified and progressive pruning for compressing vision-language transformers","venue":null,"work_id":"0eef9794-3872-481b-bfe5-777f1b437de0","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:385c0e48d95f5e74da4aa2695c48c6caea331c1c0e771b2a04095197c37e76bd","observation_id":"bb6f703d-4cee-4d26-a111-1429808c8d65","resolution":{"observed_at":"2026-05-14T22:18:04.894713Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.05417","last_updated":"2024-12-25T11:55:30Z","snapshot_observed_at":"2026-08-18T15:14:08.660328Z","submitted_at":"2024-07-07T15:44:42Z","title":"See Further for Parameter Efficient Fine-tuning by Standing on the Shoulders of Decomposition","version":2},"cited_work":{"arxiv_id":"2407.05417","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2407.05417","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"See further for parameter efficient fine-tuning by standing on the shoulders of decomposition","venue":null,"work_id":"2ecda36a-35d9-4c3d-a04c-bcb6aca58513","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"cited_paper":"/paper/2407.05417","citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:eda3103b398a95729c3fdea55e9caa3b2c949c45f24f2e9a4992ab615058c576","observation_id":"0035b8d8-4299-4d27-96cf-edc77f80ecf1","resolution":{"observed_at":"2026-05-14T22:18:04.008009Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Sparse low rank factorization for deep neural network compression.Neurocomputing, 398:185– 196","venue":null,"work_id":"049476d5-7c21-4936-8695-1130b67c61d6","year":2020},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:cf3ebf6989aa7a5e1ca322c8f170684994e264dcb8541145e433f4fd20fc06c8","observation_id":"5d9e6d6e-3842-481e-924c-f2f1181e22fb","resolution":{"observed_at":"2026-05-14T22:18:05.031289Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"9f40e77a-fccd-4cf3-bf02-770efaa8c529","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:b76033ec2d2ae57d5b1147e6bcdf18aa22bd944c8762dd42256172d5344f1a4b","observation_id":"b38ca952-3652-4215-b574-e7f5c70b7b09","resolution":{"observed_at":"2026-05-14T22:18:05.163692Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Tukan, A","venue":null,"work_id":"61ba7e75-4dd8-459f-98bb-763c83fdc8ae","year":2021},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:ddfc3d3259ccd1786d5bcd4788fe665b30020af849d024dea23226236aafb3e5","observation_id":"98290452-eb88-45b7-9dfb-990814f67b7a","resolution":{"observed_at":"2026-05-14T22:18:05.049394Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"DyLoRA: Parameter-efficient tuning of pre-trained models using dynamic search-free low-rank adaptation","venue":null,"work_id":"0f71d5cf-0d31-4ace-8ec2-567a0e5b2c0f","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":73,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:342e8b1f8aa44cb48da37b4264962893953cd45bcbc05ff754477de6274ee9a9","observation_id":"a7b0e86a-0645-413e-9368-9d96e07c5d6a","resolution":{"observed_at":"2026-05-14T22:18:05.104407Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"ec38b059-0ec8-4da4-b3fc-fb95c387e0d9","year":2011},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:e935b1d197cf04822a37f8b68fd65582684a1698ac814d3dedbd65400f3ef9e6","observation_id":"be8793e0-c720-4803-888f-0211a135442d","resolution":{"observed_at":"2026-05-14T22:18:04.822698Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"84b9398d-9380-4c06-b68c-8673757dc0f8","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:b6dddd6e245d340e433c253290c31e2bc0bc1b642f656b5648246efa5f6bcf5f","observation_id":"8ae7d75a-82a3-40b5-b7a1-046ac6ce8365","resolution":{"observed_at":"2026-05-14T22:18:04.818267Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"6535408f-8c23-47ec-af77-c620383553c0","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":76,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:1f22f748f49eb204446f42fcbbb37ed71c9afeb6a6bd0f674c32650d4dbd43ff","observation_id":"d70558d8-57a0-4518-8235-f774182d82cb","resolution":{"observed_at":"2026-05-14T22:18:04.863791Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Basis sharing: Cross-layer parameter sharing for large language model compression","venue":null,"work_id":"8bbe0f37-da47-4808-8364-94655c0c4bb5","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":77,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:bee23a84263cacc429bfc1649e44c953fab75264523dcd886054843279db0c0c","observation_id":"1775e6a1-9f4f-43fa-b4e6-77ea663404d1","resolution":{"observed_at":"2026-05-14T22:18:04.987623Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Neural network pa- rameter diffusion","venue":null,"work_id":"4a490b44-3f46-439f-a9e7-631144c5c751","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":78,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:99a8c739c5ba0edba8b5da48df3d6b1c276789ad193df4f1335a5d861bf55854","observation_id":"65fe1fcf-6e3c-4719-870b-4fe3be8e676d","resolution":{"observed_at":"2026-05-14T22:18:04.915213Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"SVD- LLM: Truncation-aware singular value decomposition for large language model compression","venue":null,"work_id":"b6225dad-b14e-4c85-b0b9-ee92e07b3361","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":79,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:d11f0b298bcfb88d51704637b70a961811a3b8d113b93065d37dd5b44352453c","observation_id":"4614fa46-7ad9-4730-86ed-cf66b1b7bdfd","resolution":{"observed_at":"2026-05-14T22:18:05.118291Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Revisiting the power of prompt for visual tuning","venue":null,"work_id":"0d636544-1648-410c-b661-0aa7fac7902c","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":80,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:ef5a9127dfb62aa98d4b912a0cbe3073ff2e5a94b84809d1da029e7fb7ea941d","observation_id":"94192cfc-b58c-4ae5-9e3c-0e3f7cbf0654","resolution":{"observed_at":"2026-05-14T22:18:04.813704Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Kd- detr: Knowledge distillation for detection transformer with consistent distillation points sampling","venue":null,"work_id":"7759f548-b1c7-454d-9867-e57ec638154a","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":81,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:a9a9caeabc5f608de0cae3fc577a805f95fe61ca1ed674d4cb8b67ec6758e5a4","observation_id":"5e6a7a86-bb03-4200-8bf9-c061810860f4","resolution":{"observed_at":"2026-05-14T22:18:04.969563Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Learning to prompt for contin- ual learning","venue":null,"work_id":"7179abd4-2966-4228-9988-44f35b7856c6","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":82,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:7616b01169b7739258f6bc122c765c538c4678a6cdae8cc4d5ef36d38a2e5e54","observation_id":"c11f5501-f739-4fb1-97e0-cbe300c04817","resolution":{"observed_at":"2026-05-14T22:18:05.059209Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Integer quantization for deep learn- ing inference: Principles and empirical evaluation","venue":null,"work_id":"107ee22d-52d7-4621-b336-193c26ffb65a","year":2020},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":83,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:f56e45600f17f59f63e62bce59cd03a4cf159b3c73a5630a087e87ea7b825547","observation_id":"b7ef24d6-acce-4bc9-b04a-ec6f7248178f","resolution":{"observed_at":"2026-05-14T22:18:04.799436Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Efficient Distortion-Minimized Layerwise Pruning .IEEE Transac- tions on Pattern Analysis & Machine Intelligence, 47(10): 9298–9315","venue":null,"work_id":"74017cc0-e1d6-4c3e-b9e3-04cc32b8ffa3","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":84,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:7623b838ec2a9ce2ed585cf78bda214e1ebc68c718d2a15683269a3cd590383f","observation_id":"c02927bd-5965-4622-97a9-fa2c29b458a1","resolution":{"observed_at":"2026-05-14T22:18:04.845868Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"LaCo: Large lan- guage model pruning via layer collapse","venue":null,"work_id":"2e6c0aab-d6b2-4c7c-8881-8b12d56a3bb0","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":85,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:230a46a32eafcebd1dff4400b7eb2b6547be6651c54345ec7c9455a06bc5a5e3","observation_id":"6667272c-fa21-4f37-8018-da4b5fcb18c8","resolution":{"observed_at":"2026-05-14T22:18:04.956280Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Vitkd: Feature-based knowledge distil- lation for vision transformers","venue":null,"work_id":"9fa73267-008f-4e71-9dae-8ad5361fa848","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":86,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:4d3f9643e0273d6cb992b349df0402a4546b8d57645d705884cbb2c862812f2c","observation_id":"612ed8f7-a756-4acd-a8d9-49def21b0bff","resolution":{"observed_at":"2026-05-14T22:18:05.000557Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2405.20985","last_updated":"2024-05-31T16:31:38Z","snapshot_observed_at":"2026-08-17T11:43:45.183126Z","submitted_at":"2024-05-31T16:31:38Z","title":"DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models","version":1},"cited_work":{"arxiv_id":"2405.20985","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2405.20985","snapshot_observed_at":"2026-07-04T16:39:57.301173Z","title":"Deco: Decoupling token compression from semantic abstraction in multimodal large language models","venue":null,"work_id":"72b8bcdd-c399-4415-a8e7-ca6f24d31b48","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":87,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"cited_paper":"/paper/2405.20985","citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:cf5371db5f72ee0a314ed4ff8d2113c61709ef1da6ce3d7a3b49f80eb07ac892","observation_id":"df90ee04-6cd9-4fb3-83bd-b921dba7fa2a","resolution":{"observed_at":"2026-05-14T22:18:04.021759Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Once for both: Single stage of importance and sparsity search for vision transformer compression","venue":null,"work_id":"8a0fb5d0-423f-4a22-b5d2-06e22a97d7a5","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":88,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:af1eabe8b949d162ee3209cfddcd082788c5c05509233a674986fa316fc9eb27","observation_id":"f1da86e8-c6c2-4239-890d-c19227f95d91","resolution":{"observed_at":"2026-05-14T22:18:04.869189Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Yu, Jianlong Chang, Haixin Wang, Lingbo Liu, Shijie Wang, Zhiyu Wang, Junfan Lin, Lingxi Xie, Haojie Li, Zhouchen Lin, Qi Tian, and Chang Wen Chen","venue":null,"work_id":"6132fcea-b923-474f-8d10-8a4a6eddf667","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":89,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:1f5571fbe644cba5deaa62549a14575fc56ae6546c1aaba815324e2920b8986b","observation_id":"ef910760-45ef-4dfa-8c56-a6aebffeeef4","resolution":{"observed_at":"2026-05-14T22:18:04.920207Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Yu and J","venue":null,"work_id":"6600a24c-ce6e-4159-91a8-0118952f614f","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":90,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:e680522da4c7fcbaad2f8f913afd9b268da893969a46410c99be78f85f96b4b6","observation_id":"d636d0c7-1315-4aa7-88ae-039673e45d14","resolution":{"observed_at":"2026-05-14T22:18:04.827210Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Hellaswag: Can a machine really finish your sentence? InProceedings of the 57th Annual Meet- ing of the Association for Computational Linguistics, pages 4791–4800","venue":null,"work_id":"c580dcee-75ec-480b-93ef-d963e522a23b","year":2019},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":91,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:df487de8435da915f1ae5acd663f576e0088538a22b4943775b9286a26440224","observation_id":"6ae55e2b-179b-4862-bba8-0408ef42e0f3","resolution":{"observed_at":"2026-05-14T22:18:04.964587Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Vi- sual fourier prompt tuning","venue":null,"work_id":"0ed77301-a6cb-4c4c-8f29-8bddeb11b6bb","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":92,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:024da803726d1e8ce6f2ef15e5e90e1b3a8fffa11fd84c04e8c6c98756ced1a9","observation_id":"48b0fdf7-7bc3-4c6d-ae1c-cd0d6164b7f2","resolution":{"observed_at":"2026-05-14T22:18:04.983717Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"A large-scale study of represen- tation learning with the visual task adaptation benchmark","venue":null,"work_id":"2ba03901-faf4-4bf5-bc69-f7ce321c5e4e","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":93,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:9de5075cd9197e68b33f47428afe0d537e920597351081163404e26f6d0dc8d8","observation_id":"7fdaf10f-a96e-4726-a54a-090941389422","resolution":{"observed_at":"2026-05-14T22:18:05.036040Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Metadiff: Meta- learning with conditional diffusion for few-shot learning","venue":null,"work_id":"6dd1466b-cb38-4498-a031-6813fe7ecbdd","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":94,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:06ec107aa7e62d9e616051c174fe4f082295a337a92527f2279842b07490c826","observation_id":"a9b2a52b-5b55-4e9f-9b24-d380ec27aac6","resolution":{"observed_at":"2026-05-14T22:18:05.173522Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Dense vision transformer compression with few samples","venue":null,"work_id":"227e17bc-447e-4ead-854f-081d715561bb","year":null},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":95,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:d3c76b300db6ed74b9719b19a3fb12e542ef7f6ec9f201893c23008ff972be06","observation_id":"028b333c-aa4a-485d-a548-ee9c844d3126","resolution":{"observed_at":"2026-05-14T22:18:05.092823Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Minivit: Compressing vi- sion transformers with weight multiplexing","venue":null,"work_id":"3553215e-eda2-46d5-bfb5-97ee8270e6ad","year":2022},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":96,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:0f831a4445a3472bf2017dd78caf64494d6b39c0e6a1192da2e3c33eeaf473b2","observation_id":"8c8dde77-e163-42b4-ba8d-cbd4fbceb34e","resolution":{"observed_at":"2026-05-14T22:18:05.145770Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Zhang, X","venue":null,"work_id":"322cc532-1dd4-4e13-9c82-9b85fafb8681","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":97,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:ab22ba5b0ddba04afed1147ddaefc562a64bcfdfe29627cb63897c077279d7cd","observation_id":"0c96fb7f-5b91-4c97-bae8-86f48ab8719f","resolution":{"observed_at":"2026-05-14T22:18:04.951504Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"LoRAPrune: Structured pruning meets low-rank parameter-efficient fine- tuning","venue":null,"work_id":"9d32a7c2-3711-4905-9d8e-752bec0b4ba7","year":2024},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":98,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:481e097da2a95040670928de9296b96d5821f7ece0176ba3b9527934d1f05085","observation_id":"1dbdf622-ff0c-4db6-b04d-d3960cbf8a69","resolution":{"observed_at":"2026-05-14T22:18:04.883173Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Intra-head pruning for vision transformers via inter-layer dimension relationship modeling.Neural Networks, 190: 107656","venue":null,"work_id":"f1f8ce8b-111a-4bfb-8720-54a20b92beb4","year":2025},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":99,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:1bdff24d255881e984053a70602a63e7dcdc068b8d419c9ef64c5b63f008a43b","observation_id":"4d871a7f-f28b-488e-b65c-bde5f12174cb","resolution":{"observed_at":"2026-05-14T22:18:05.122568Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Adaptive budget allocation for parameter-efficient fine- tuning","venue":null,"work_id":"0e349897-2ee6-4151-af62-a539c1fb5eca","year":2023},"citing_paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression","version":2},"reference_index":100,"source":"pdf_text","source_observed_at":"2026-05-14T22:17:47.203589Z"},"links":{"citing_paper":"/paper/2603.27383"},"observation_digest":"sha256:b90cdb44f28b3a65c58964ee101f4b08de334f0fc2fbac7e1b7e23c4e32300c2","observation_id":"37294b73-4ffb-4407-ae35-e63a2f6fd513","resolution":{"observed_at":"2026-05-14T22:18:05.159342Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2603.27383","last_updated":"2026-04-10T19:04:13Z","latest_version":2,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-11T14:39:01.939262Z","submitted_at":"2026-03-28T19:29:38Z","title":"Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression"},"reference_resolution":{"displayed":100,"state_counts":{"malformed_identifier":1,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":6,"verified_exact":5,"verified_fuzzy":88},"total_outbound_references":102},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"thesis":"As of 18 August 2026, this Paper Citation Record lists 100 of 102 outbound references and 0 inbound Pith citation observations for arXiv:2603.27383."}