{"as_of":"2026-08-21T00:15:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:a67e94f940dc09487498d4f0b3ddd8e1f8b2b4e6379af619c4a692ca4c4997f9","coverage":[{"denominator":37,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":37,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T19:04:16.666931Z","state":"measured"},{"denominator":39,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":39,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-20T06:33:59.587034+00:00","state":"measured"},{"denominator":2,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":2,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T05:18:28.864015Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-05-10T06:16:20.984883Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.17781","snapshot_observed_at":"2026-08-07T05:18:28.864015Z","title":"D., Dinu, G., and Karypis, G","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.08354","last_updated":"2026-05-28T08:13:47Z","snapshot_observed_at":"2026-08-10T06:36:57.892160Z","submitted_at":"2025-06-10T02:11:42Z","title":"Position: Text Embeddings Should Capture Implicit Semantics, Not Just Surface Meaning","version":2},"reference_index":86,"source":"pdf_text","source_observed_at":"2026-08-07T05:18:28.864015Z"},"links":{"cited_paper":"/paper/2506.17781","citing_paper":"/paper/2506.08354"},"observation_digest":"sha256:e3db8ef1f126a9a319a74dedcf4df5026c02f732a1888e544e7a0fd708b87977","observation_id":"87b462eb-8bd6-4926-afd0-7df29d3c4777","resolution":{"observed_at":"2026-08-07T05:18:28.864015Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"cited_work":{"arxiv_id":"2506.17781","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2506.17781","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"53f46308-9da2-4ba0-b4fd-a14e3bc32f6a","year":2025},"citing_paper":{"arxiv_id":"2604.17257","last_updated":"2026-04-21T06:47:04Z","snapshot_observed_at":"2026-08-11T09:18:25.699659Z","submitted_at":"2026-04-19T04:41:55Z","title":"REZE: Representation Regularization for Domain-adaptive Text Embedding Pre-finetuning","version":2},"reference_index":59,"source":"arxiv_source","source_observed_at":"2026-05-10T06:14:28.893342Z"},"links":{"cited_paper":"/paper/2506.17781","citing_paper":"/paper/2604.17257"},"observation_digest":"sha256:117b2c9d019ab2e017132881334b7a5497c0b44b1f7c107e26d13f3367366391","observation_id":"d267bc57-a095-484e-b70b-527100b79053","resolution":{"observed_at":"2026-05-10T06:16:20.986162Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2506.17781/citation-record","integrity":"/paper/2506.17781/integrity","json":"/paper/2506.17781/citation-record.json","paper":"/paper/2506.17781"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.511422Z","title":null,"venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.511422Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:ed9d7607b5aa17c02a7b0cb6271c852c8ffef0e768031bd51ebd8b565ef35cf2","observation_id":"8ea532a5-6f99-4634-9b9b-9a21c9d5a46d","resolution":{"observed_at":"2026-08-15T19:04:16.511422Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1607.06450","last_updated":"2016-07-21T19:57:52Z","snapshot_observed_at":"2026-08-15T04:53:45.483331Z","submitted_at":"2016-07-21T19:57:52Z","title":"Layer Normalization","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1607.06450","snapshot_observed_at":"2026-08-15T19:04:16.515676Z","title":null,"venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.515676Z"},"links":{"cited_paper":"/paper/1607.06450","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:176ab707961746582567228ce970fea0b6801982196beec145b2e1b4b8542714","observation_id":"4753733e-7900-4321-b650-674467fdb60b","resolution":{"observed_at":"2026-08-15T19:04:16.515676Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.520065Z","title":null,"venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.520065Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:7ad424aa1237dde269e242affeba8da4cfa1b0852b0247ba9412cbd6856f34c5","observation_id":"bbc529d0-0d41-4fbb-97b5-908c60bcb782","resolution":{"observed_at":"2026-08-15T19:04:16.520065Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:17.291215Z","title":null,"venue":null,"work_id":"08a96816-9ba3-473c-bcc2-7ce6e57e482d","year":2021},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.524882Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:12971f0fcd1d535edd4cfaae8bcdd3204545970cdceba5fd2680bfafec79abd7","observation_id":"bb52473c-6484-44d8-af05-a58454235a3c","resolution":{"observed_at":"2026-08-15T19:04:17.295681Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1810.04805","last_updated":"2019-05-24T20:37:26Z","snapshot_observed_at":"2026-08-14T18:16:28.847993Z","submitted_at":"2018-10-11T00:50:01Z","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1810.04805","snapshot_observed_at":"2026-08-15T19:04:16.529287Z","title":null,"venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.529287Z"},"links":{"cited_paper":"/paper/1810.04805","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:368309afef92cb5d496cbe531d70d7e0a9bef89e0f35e302eeeab9213dbfdb0f","observation_id":"0fe9c641-22f3-4b6c-ae71-242e74f07dcf","resolution":{"observed_at":"2026-08-15T19:04:16.529287Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.534266Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.534266Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:f2334e8744813b98eaf12feacf24934eedf5a5ccc887deb6c7d7c8aea0ccf6fb","observation_id":"d6b6fecd-81ef-4d06-af0d-2d0dfbbbf70e","resolution":{"observed_at":"2026-08-15T19:04:16.534266Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2104.08821","last_updated":"2022-05-18T12:29:49Z","snapshot_observed_at":"2026-07-06T11:01:05.577957Z","submitted_at":"2021-04-18T11:27:08Z","title":"SimCSE: Simple Contrastive Learning of Sentence Embeddings","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2104.08821","snapshot_observed_at":"2026-08-15T19:04:16.538886Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.538886Z"},"links":{"cited_paper":"/paper/2104.08821","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:aa766b89d21fbb4dfca936c73b4396cecc1a07a71dfe3e654122811363abe13c","observation_id":"b1594f96-1861-4af1-b00d-ef89020cc435","resolution":{"observed_at":"2026-08-15T19:04:16.538886Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2312.10997","last_updated":"2024-03-27T09:16:57Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-12-18T07:47:33Z","title":"Retrieval-Augmented Generation for Large Language Models: A Survey","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.10997","snapshot_observed_at":"2026-08-15T19:04:16.543466Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.543466Z"},"links":{"cited_paper":"/paper/2312.10997","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:e667f5c0e71a0a7eddba8d4ab84dc010b4555f1fa17e69e1ea4da4315cdff7ed","observation_id":"d1effff1-a312-42c4-a746-1c4477ac2f63","resolution":{"observed_at":"2026-08-15T19:04:16.543466Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2203.01104","last_updated":"2022-10-10T06:41:52Z","snapshot_observed_at":"2026-08-16T17:17:31.222996Z","submitted_at":"2022-03-02T13:44:49Z","title":"Parameter-Efficient Mixture-of-Experts Architecture for Pre-trained Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2203.01104","snapshot_observed_at":"2026-08-15T19:04:16.547549Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.547549Z"},"links":{"cited_paper":"/paper/2203.01104","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:7ebb3efdefacf65acc7868f202933ac1563e70651197f145a5d1dc2e4142da4b","observation_id":"1312e39f-1b42-4200-9afa-5f4ea69f4a0f","resolution":{"observed_at":"2026-08-15T19:04:16.547549Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1016/b978-0-12-381479-1.00002-2","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.704464Z","title":null,"venue":null,"work_id":"b82f92b6-9e34-4caf-a5a4-22822cd8e768","year":2012},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.551971Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:a94a69c425e4f1ce4233eac9c7bbaead730fb7a9dd03d9379b3db836f46b9a2d","observation_id":"1740b463-56e5-4501-a7ab-8f6fa3a50da2","resolution":{"observed_at":"2026-08-15T19:04:16.709309Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.07524","last_updated":"2025-06-16T02:29:18Z","snapshot_observed_at":"2026-08-16T13:10:53.411712Z","submitted_at":"2024-10-10T01:36:03Z","title":"Upcycling Large Language Models into Mixture of Experts","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.07524","snapshot_observed_at":"2026-08-15T19:04:16.556019Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.556019Z"},"links":{"cited_paper":"/paper/2410.07524","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:367d3722f59aa0ac1224e119bde4576921250a56f2cacb9b191e814ccbfe1560","observation_id":"169c5b97-1b61-47e1-a521-63ef42a88e7d","resolution":{"observed_at":"2026-08-15T19:04:16.556019Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.04088","last_updated":"2024-01-08T18:47:34Z","snapshot_observed_at":"2026-08-13T19:43:49.936776Z","submitted_at":"2024-01-08T18:47:34Z","title":"Mixtral of Experts","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.04088","snapshot_observed_at":"2026-08-15T19:04:16.560636Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.560636Z"},"links":{"cited_paper":"/paper/2401.04088","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:c7a7195b796b2706ed9c499fadc8686509a7de7b395af28c034f32472a3a6408","observation_id":"a478797b-96a2-4096-9f69-349156bac4e1","resolution":{"observed_at":"2026-08-15T19:04:16.560636Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2212.05055","last_updated":"2023-02-17T17:54:50Z","snapshot_observed_at":"2026-08-20T19:56:14.362732Z","submitted_at":"2022-12-09T18:57:37Z","title":"Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.05055","snapshot_observed_at":"2026-08-15T19:04:16.565265Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.565265Z"},"links":{"cited_paper":"/paper/2212.05055","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:ed29f16666a8f7dea56c4ab4ffbc671e70dfcc10240fa2d0879bc466a942b309","observation_id":"d5dcdc78-c5ab-4025-9050-0c6f9e1cc6f1","resolution":{"observed_at":"2026-08-15T19:04:16.565265Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2006.16668","last_updated":"2020-06-30T10:42:02Z","snapshot_observed_at":"2026-08-07T09:27:36.420559Z","submitted_at":"2020-06-30T10:42:02Z","title":"GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2006.16668","snapshot_observed_at":"2026-08-15T19:04:16.569568Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.569568Z"},"links":{"cited_paper":"/paper/2006.16668","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:68e853a95e653ce9b699cd04966e12bdb4ecb43ddd5b8a8a2eb53d295e913fff","observation_id":"84ec9c00-a74d-4426-b91a-b1880c9f8329","resolution":{"observed_at":"2026-08-15T19:04:16.569568Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.573801Z","title":"u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt \\","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.573801Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:89734bd55f96cf21ed12d25cb33c8c699c3db7ac7c1775ae82a7d6ecdf04e6cb","observation_id":"4ed120ab-ee2b-47f2-a832-c48dfc808c12","resolution":{"observed_at":"2026-08-15T19:04:16.573801Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1907.11692","last_updated":"2019-07-26T17:48:29Z","snapshot_observed_at":"2026-08-16T14:33:50.657682Z","submitted_at":"2019-07-26T17:48:29Z","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1907.11692","snapshot_observed_at":"2026-08-15T19:04:16.577827Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.577827Z"},"links":{"cited_paper":"/paper/1907.11692","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:e8cfeb87691b45a15b2ba6af19bd14151b8998eb83550a5e095d357a52d899fe","observation_id":"be9bd6ac-f0fe-4891-b8a2-b4f1136996f1","resolution":{"observed_at":"2026-08-15T19:04:16.577827Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.09906","last_updated":"2025-03-03T04:28:49Z","snapshot_observed_at":"2026-08-17T08:29:24.619960Z","submitted_at":"2024-02-15T12:12:19Z","title":"Generative Representational Instruction Tuning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.09906","snapshot_observed_at":"2026-08-15T19:04:16.582101Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.582101Z"},"links":{"cited_paper":"/paper/2402.09906","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:e9c0c2f9275d8eaf30c44b06491621ac3d378712e519a8f51451e6355e28f366","observation_id":"99cb1156-5dac-48dd-abed-75ed9553b0ef","resolution":{"observed_at":"2026-08-15T19:04:16.582101Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2210.07316","last_updated":"2023-03-19T13:37:01Z","snapshot_observed_at":"2026-08-17T06:54:11.300249Z","submitted_at":"2022-10-13T19:42:08Z","title":"MTEB: Massive Text Embedding Benchmark","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.07316","snapshot_observed_at":"2026-08-15T19:04:16.586413Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.586413Z"},"links":{"cited_paper":"/paper/2210.07316","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:7636f88929b49461de8cd8bdb14642ca176acfa3cbd84aeccdb419308849e776","observation_id":"c9484877-f797-4019-b783-6b57c64e6a8d","resolution":{"observed_at":"2026-08-15T19:04:16.586413Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2201.10005","last_updated":"2022-01-24T23:36:20Z","snapshot_observed_at":"2026-07-06T12:30:51.934079Z","submitted_at":"2022-01-24T23:36:20Z","title":"Text and Code Embeddings by Contrastive Pre-Training","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.10005","snapshot_observed_at":"2026-08-15T19:04:16.590563Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.590563Z"},"links":{"cited_paper":"/paper/2201.10005","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:cec7b70388c5111858ec2098c280d88c1d0b05d1d5dc50d7c844f31f0160c7fa","observation_id":"e5fcfaab-0ba4-4943-bae0-54f2afc2af10","resolution":{"observed_at":"2026-08-15T19:04:16.590563Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.01613","last_updated":"2025-02-03T22:26:56Z","snapshot_observed_at":"2026-08-14T15:03:32.894363Z","submitted_at":"2024-02-02T18:23:18Z","title":"Nomic Embed: Training a Reproducible Long Context Text Embedder","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.01613","snapshot_observed_at":"2026-08-15T19:04:16.595386Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.595386Z"},"links":{"cited_paper":"/paper/2402.01613","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:6f9754329ef5a520cb012ad9eabeaa8e84bd528507143a031c1e8093b25230d1","observation_id":"48984a02-610f-44f6-a6c5-87059c8f454d","resolution":{"observed_at":"2026-08-15T19:04:16.595386Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.600381Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.600381Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:74c57ac71d240f7e711d7402cc27390b2103d53747919373718e489f236e1ed7","observation_id":"c0f37475-2620-47f7-8648-cac033056891","resolution":{"observed_at":"2026-08-15T19:04:16.600381Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1807.03748","last_updated":"2019-01-22T18:47:12Z","snapshot_observed_at":"2026-08-14T18:53:38.574749Z","submitted_at":"2018-07-10T16:52:11Z","title":"Representation Learning with Contrastive Predictive Coding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1807.03748","snapshot_observed_at":"2026-08-15T19:04:16.604497Z","title":null,"venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.604497Z"},"links":{"cited_paper":"/paper/1807.03748","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:bf410e518fa0ea7a493f6878a3d6b4d1099f595f2d33e6438af814183d61626c","observation_id":"103c5f08-4f8b-415e-897c-b59acb181ee6","resolution":{"observed_at":"2026-08-15T19:04:16.604497Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2007.10527","last_updated":"2021-01-05T18:16:35Z","snapshot_observed_at":"2026-08-17T17:18:44.603084Z","submitted_at":"2020-07-20T23:26:16Z","title":"Navigating the Trade-Off between Multi-Task Learning and Learning to Multitask in Deep Neural Networks","version":2},"cited_work":{"arxiv_id":"2007.10527","doi":null,"metadata_source":"pith","pith_arxiv_id":"2007.10527","snapshot_observed_at":"2026-08-15T19:04:16.953283Z","title":"Navigating the Trade-Off between Multi-Task Learning and Learning to Multitask in Deep Neural Networks","venue":"cs.LG","work_id":"6c11cd38-67e7-43d7-911c-b84de3fba8be","year":2020},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.608591Z"},"links":{"cited_paper":"/paper/2007.10527","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:7effcc46c0a5f776f964d5495d9ae21a23c3d9cf68b741448c0c6f6474aea1b2","observation_id":"114d2067-af20-480e-888c-2ccd5098632f","resolution":{"observed_at":"2026-08-15T19:04:16.958270Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.612633Z","title":null,"venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.612633Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:0d8ef75d8dbebe0aceb69271f54c7acc2fe2e75298ea723c67d9342df4895bf6","observation_id":"6670f640-e5a4-4183-9ca3-80af65013b1d","resolution":{"observed_at":"2026-08-15T19:04:16.612633Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2212.09741","last_updated":"2023-05-30T15:22:50Z","snapshot_observed_at":"2026-08-19T13:00:26.130369Z","submitted_at":"2022-12-19T18:57:05Z","title":"One Embedder, Any Task: Instruction-Finetuned Text Embeddings","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.09741","snapshot_observed_at":"2026-08-15T19:04:16.616530Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.616530Z"},"links":{"cited_paper":"/paper/2212.09741","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:77c81ec560cb298c60b0a6c5c2813a32f109a4868736d27c31f9abee6193b8c5","observation_id":"93fdc0ae-c6b8-4c5e-b550-4a3f12548709","resolution":{"observed_at":"2026-08-15T19:04:16.616530Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2411.14708","last_updated":"2025-02-15T16:26:55Z","snapshot_observed_at":"2026-08-18T23:09:49.081094Z","submitted_at":"2024-11-22T03:33:51Z","title":"Understanding LLM Embeddings for Regression","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2411.14708","snapshot_observed_at":"2026-08-15T19:04:16.621321Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.621321Z"},"links":{"cited_paper":"/paper/2411.14708","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:bfdc17956f4f315efb6746aa4f82f66ed8e547373bfd44a16d6dbf677e006f21","observation_id":"3e4b247e-f54f-4d6b-88ab-ed5ca114cdfc","resolution":{"observed_at":"2026-08-15T19:04:16.621321Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"physics/0004057","last_updated":"2000-04-24T15:22:30Z","snapshot_observed_at":"2026-08-16T09:10:10.815975Z","submitted_at":"2000-04-24T15:22:30Z","title":"The information bottleneck method","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"physics/0004057","snapshot_observed_at":"2026-08-15T19:04:16.625478Z","title":null,"venue":null,"work_id":null,"year":2000},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.625478Z"},"links":{"cited_paper":"/paper/physics/0004057","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:7ab61bb300dc505e2ddfe6d690ee876dc2122f708b7d4c0ae602ea62863282fb","observation_id":"431eda90-0064-4b1e-84d9-4db53301e06b","resolution":{"observed_at":"2026-08-15T19:04:16.625478Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2212.03533","last_updated":"2024-02-22T06:21:51Z","snapshot_observed_at":"2026-07-06T14:27:46.217000Z","submitted_at":"2022-12-07T09:25:54Z","title":"Text Embeddings by Weakly-Supervised Contrastive Pre-training","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.03533","snapshot_observed_at":"2026-08-15T19:04:16.630076Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.630076Z"},"links":{"cited_paper":"/paper/2212.03533","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:68386d8f3ce1fc93a5e20a90b5e34c088fd9ec201ffa02d39178d6c650aaa4ef","observation_id":"b4e8a1d6-f50c-4dda-82f9-b303f83e4eef","resolution":{"observed_at":"2026-08-15T19:04:16.630076Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.00368","last_updated":"2024-05-31T07:22:01Z","snapshot_observed_at":"2026-08-20T00:24:38.555810Z","submitted_at":"2023-12-31T02:13:18Z","title":"Improving Text Embeddings with Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.00368","snapshot_observed_at":"2026-08-15T19:04:16.634581Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.634581Z"},"links":{"cited_paper":"/paper/2401.00368","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:528173282ddffa1ddb33e7b99d95ae5e2fcddca6327f21862eda2d0ddc4c724a","observation_id":"c022c87b-a518-4d5b-a1de-d435c5b35115","resolution":{"observed_at":"2026-08-15T19:04:16.634581Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.05672","last_updated":"2024-02-08T13:47:50Z","snapshot_observed_at":"2026-08-12T15:58:37.148545Z","submitted_at":"2024-02-08T13:47:50Z","title":"Multilingual E5 Text Embeddings: A Technical Report","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.05672","snapshot_observed_at":"2026-08-15T19:04:16.638816Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.638816Z"},"links":{"cited_paper":"/paper/2402.05672","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:b6adb054da24eb02ec2630bb755e610d745e007391193b7201d5745270ef178b","observation_id":"155f2d82-08f4-408c-befc-228f76f9d5de","resolution":{"observed_at":"2026-08-15T19:04:16.638816Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.643184Z","title":null,"venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.643184Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:2636e5f153d57c5840e0ccace0092ef25db074440e3f81759f2f48e7119ef6ce","observation_id":"cca53b07-7e74-4714-8b26-b0fb928a4387","resolution":{"observed_at":"2026-08-15T19:04:16.643184Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:17.249727Z","title":null,"venue":null,"work_id":"fb15058b-b525-4006-aebe-f7ed5dbf5c99","year":2021},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.647128Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:2d986f6338afd32e7f74441e884e5cd75d870dbeda2e5c0be10ee346424bb943","observation_id":"16fdde60-9b68-41a7-ac61-dc06271d4a4e","resolution":{"observed_at":"2026-08-15T19:04:17.254019Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2007.12731","last_updated":"2020-07-24T18:29:43Z","snapshot_observed_at":"2026-08-10T20:35:28.844168Z","submitted_at":"2020-07-24T18:29:43Z","title":"COVID-19 Knowledge Graph: Accelerating Information Retrieval and Discovery for Scientific Literature","version":1},"cited_work":{"arxiv_id":"2007.12731","doi":null,"metadata_source":"pith","pith_arxiv_id":"2007.12731","snapshot_observed_at":"2026-08-15T19:04:16.869883Z","title":"COVID-19 Knowledge Graph: Accelerating Information Retrieval and Discovery for Scientific Literature","venue":"cs.IR","work_id":"b880e394-86ee-41ba-b4ba-8b59352132aa","year":2020},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.651166Z"},"links":{"cited_paper":"/paper/2007.12731","citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:fed5689cf28bedca9a1451ad7c7045fc3ba9063827a49ba6ff760e173e2c5313","observation_id":"cf89cc58-8960-41c0-9346-09098c8a073c","resolution":{"observed_at":"2026-08-15T19:04:16.875192Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.655353Z","title":null,"venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.655353Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:8df12dcf04d42c81eccef4065d6ac7c4ceda178c939173c6218a3557437e6c98","observation_id":"9fbeef19-8a33-469c-9f4a-02906850a957","resolution":{"observed_at":"2026-08-15T19:04:16.655353Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.659088Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.659088Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:6bbe0fa3ed5c9b4ad55e2aa9e4da2f5a61629fa0f141920229f0cab5021e79d3","observation_id":"eb4d1faf-76fc-41e4-891e-8f4584374ff9","resolution":{"observed_at":"2026-08-15T19:04:16.659088Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:17.221442Z","title":null,"venue":null,"work_id":"6c2186af-0485-4806-bd23-c43c3627d08a","year":2012},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.663121Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:9f97057a68cae8a0b39a15911e3df4ac1ae847406f10b6c92c57e0e738a42ef5","observation_id":"a9e7bb63-be4d-40f3-8c76-5445833863af","resolution":{"observed_at":"2026-08-15T19:04:17.226076Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:04:16.666931Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models","version":1},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-08-15T19:04:16.666931Z"},"links":{"citing_paper":"/paper/2506.17781"},"observation_digest":"sha256:ed837e1825976e38d7494f0f4b06e1ceb8384eda603f42a2ae664ed26f3d6050","observation_id":"d438bcb3-7c3c-45ae-8db5-78ece9a241b8","resolution":{"observed_at":"2026-08-15T19:04:16.666931Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2506.17781","last_updated":"2025-06-21T18:28:25Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-18T09:56:03.055494Z","submitted_at":"2025-06-21T18:28:25Z","title":"Beyond instruction-conditioning, MoTE: Mixture of Task Experts for Multi-task Embedding Models"},"reference_resolution":{"displayed":37,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":34,"verified_exact":3,"verified_fuzzy":0},"total_outbound_references":37},"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":"2026-08-20T06:33:59.587034+00:00","source":"crossref"},{"observed_at":"2026-08-20T06:33:54.927442+00:00","source":"retraction_watch"}],"thesis":"As of 21 August 2026, this Paper Citation Record lists 37 of 37 outbound references and 2 inbound Pith citation observations for arXiv:2506.17781."}