{"as_of":"2026-08-17T22:12:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:15fff19039244da9214a2596b5f5062f99a99176909538a710fb4393a34c2566","coverage":[{"denominator":61,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":61,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T20:36:40.017952Z","state":"measured"},{"denominator":63,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":63,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-17T06:30:58.91139+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":"2026-08-12T15:52:45.940026Z","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":"2026-05-12T07:31:37.973339Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"cited_work":{"arxiv_id":"2505.12512","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2505.12512","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Scalable strategies for continual learning with replay","venue":null,"work_id":"08e72142-2e21-4210-806b-d643c6e4ab02","year":2025},"citing_paper":{"arxiv_id":"2605.09608","last_updated":"2026-05-10T15:40:44Z","snapshot_observed_at":"2026-08-17T14:09:47.734956Z","submitted_at":"2026-05-10T15:40:44Z","title":"Geometry Conflict: Explaining and Controlling Forgetting in LLM Continual Post-Training","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-05-12T02:32:08.672114Z"},"links":{"cited_paper":"/paper/2505.12512","citing_paper":"/paper/2605.09608"},"observation_digest":"sha256:e88982e520b900c898870817f1db8690777745ee710795444bd886f344aa0c1f","observation_id":"2f44ee6f-2ef0-49ec-a217-d0177ddb041c","resolution":{"observed_at":"2026-05-12T07:31:38.010166Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.12512","snapshot_observed_at":"2026-08-12T15:52:45.940026Z","title":"arXiv preprint arXiv:2505.12512 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.10850","last_updated":"2026-08-11T12:23:28Z","snapshot_observed_at":"2026-08-16T23:59:06.685576Z","submitted_at":"2026-08-11T12:23:28Z","title":"Diffract: Spectral View of LLM Domain Adaptation","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-12T15:52:45.940026Z"},"links":{"cited_paper":"/paper/2505.12512","citing_paper":"/paper/2608.10850"},"observation_digest":"sha256:83970e2286dccc52864769adf0463b453fde5c693ecd2a06848cc63d4e745a46","observation_id":"1424c1d9-c3e7-409b-9289-51a06d2c05a3","resolution":{"observed_at":"2026-08-12T15:52:45.940026Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2505.12512/citation-record","integrity":"/paper/2505.12512/integrity","json":"/paper/2505.12512/citation-record.json","paper":"/paper/2505.12512"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T20:36:41.158483Z","title":"Ss-il: Separated softmax for incremental learning","venue":null,"work_id":"3cf22804-b94e-4884-a77d-31c35e6d7abd","year":2021},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.685795Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:c841f161b4d691b89ee85d007ceedd7578c8cfc9c4b5d1cbcda18440538fe40c","observation_id":"985f05ce-91b7-4ef2-8f84-84997dd1a033","resolution":{"observed_at":"2026-08-15T20:36:41.163084Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2502.08606","last_updated":"2025-07-25T16:55:43Z","snapshot_observed_at":"2026-08-13T20:59:35.843786Z","submitted_at":"2025-02-12T17:52:47Z","title":"Distillation Scaling Laws","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.08606","snapshot_observed_at":"2026-08-15T20:36:39.691481Z","title":"Distillation scaling laws.arXiv preprint arXiv:2502.08606, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.691481Z"},"links":{"cited_paper":"/paper/2502.08606","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:b1b084a6adf858f45840d3b5b507b1f58891da70b83ee70107f1c3f93c2b9390","observation_id":"7471c7ee-c21e-405e-b589-346dcdc6b39c","resolution":{"observed_at":"2026-08-15T20:36:39.691481Z","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":"2026-08-15T20:36:39.696283Z","title":"Dark experience for general continual learning: a strong, simple baseline.Advances in Neural Information Processing Systems, 33:15920–15930, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.696283Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:d207998ffcb6c4f4374004f2e435ccd01adba71d0b2aeb0ad3d861f544a7065d","observation_id":"6e9377d6-5dd5-4767-b7da-1084cac78dad","resolution":{"observed_at":"2026-08-15T20:36:39.696283Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.00234","last_updated":"2024-10-05T11:47:02Z","snapshot_observed_at":"2026-07-06T14:36:25.690733Z","submitted_at":"2022-12-31T15:57:09Z","title":"A Survey on In-context Learning","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.00234","snapshot_observed_at":"2026-08-15T20:36:39.701007Z","title":"A survey on in-context learning.arXiv preprint arXiv:2301.00234, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.701007Z"},"links":{"cited_paper":"/paper/2301.00234","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:5ccd828b8817a6aff411081b54e6df4599f0465fe48fb1b39c0058cdcb93806e","observation_id":"788d1a61-bef7-4ca7-b183-ddda899cfdb8","resolution":{"observed_at":"2026-08-15T20:36:39.701007Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.21783","last_updated":"2024-11-23T23:27:33Z","snapshot_observed_at":"2026-08-13T17:20:44.002518Z","submitted_at":"2024-07-31T17:54:27Z","title":"The Llama 3 Herd of Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.21783","snapshot_observed_at":"2026-08-15T20:36:39.705276Z","title":"The llama 3 herd of models.arXiv preprint arXiv:2407.21783, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.705276Z"},"links":{"cited_paper":"/paper/2407.21783","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:a4b4198d3b391cf9815c9fdb00f0b5d3a14dac62aa8f590fb72dec5bb6e0ba3d","observation_id":"bd8ccb5d-a93a-4858-9a25-4ef4e49d2a08","resolution":{"observed_at":"2026-08-15T20:36:39.705276Z","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":"2026-08-15T20:36:41.135554Z","title":"A unified continual learn- ing framework with general parameter-efficient tuning","venue":null,"work_id":"fb183c39-cf44-4d04-a7c1-a0767d12d100","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.709463Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:e177085e329b73b3b3449aee439f4f0bf7b9b3b8cca6bd8579c4c8f2dba7e3ff","observation_id":"e1f1c890-244a-4890-bbb2-2137017cc03d","resolution":{"observed_at":"2026-08-15T20:36:41.139931Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.16226","last_updated":"2024-03-21T04:47:27Z","snapshot_observed_at":"2026-08-17T04:21:00.630374Z","submitted_at":"2023-10-24T22:41:14Z","title":"TiC-CLIP: Continual Training of CLIP Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.16226","snapshot_observed_at":"2026-08-15T20:36:39.713990Z","title":"Tic-clip: Continual training of clip models.arXiv preprint arXiv:2310.16226, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.713990Z"},"links":{"cited_paper":"/paper/2310.16226","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:8b7bec7ab517dfb175e356412be7949d50e71dcf89ae48e49855461fe4e634f8","observation_id":"859c610a-853a-4703-abb8-f362e8785436","resolution":{"observed_at":"2026-08-15T20:36:39.713990Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.12948","last_updated":"2026-01-04T03:57:36Z","snapshot_observed_at":"2026-08-15T12:33:55.451951Z","submitted_at":"2025-01-22T15:19:35Z","title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.12948","snapshot_observed_at":"2026-08-15T20:36:39.719480Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.719480Z"},"links":{"cited_paper":"/paper/2501.12948","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:b357cb38e5eb067d06e268a1c1b966671b0324523b7e1d876c173db27378562d","observation_id":"3d9e1a87-df4d-4b01-bc48-73f02e670776","resolution":{"observed_at":"2026-08-15T20:36:39.719480Z","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":"2026-08-15T20:36:41.120368Z","title":"Rusu, and Razvan Pascanu","venue":null,"work_id":"cb09cac8-d7e3-42dd-bd9d-e57792cee37d","year":2020},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.725893Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:bed898e145a3b1bddd11f5c0f31e6cb66587d38751b13655c6e19cef76897796","observation_id":"6f80fae7-e518-4f9f-bf7d-33a0f627933e","resolution":{"observed_at":"2026-08-15T20:36:41.125192Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.14608","last_updated":"2024-09-16T02:54:50Z","snapshot_observed_at":"2026-08-04T09:07:42.158421Z","submitted_at":"2024-03-21T17:55:50Z","title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.14608","snapshot_observed_at":"2026-08-15T20:36:39.730477Z","title":"Parameter-efficient fine-tuning for large models: A comprehensive survey.arXiv preprint arXiv:2403.14608, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.730477Z"},"links":{"cited_paper":"/paper/2403.14608","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:b8c43b5c09c6b4f1429b5a1520c1b9c1ede89a775bfe8af6e80615ed26e46818","observation_id":"3870a8a1-62b1-453f-b26d-1e10f0a3636b","resolution":{"observed_at":"2026-08-15T20:36:39.730477Z","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":"2026-08-15T20:36:41.106784Z","title":"Hayes, Ronald Kemker, and Christopher Kanan","venue":null,"work_id":"080b8136-655b-42cb-8b9e-4e69150c30d3","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.736161Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:b85d671a9ae9404dfac5e5ceaedc881c3f72a8609bb833c4c479fa047a4745ea","observation_id":"f922df78-30c3-4c7a-b258-5f3f7f8d154b","resolution":{"observed_at":"2026-08-15T20:36:41.111002Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2104.04132","last_updated":"2021-05-28T21:01:25Z","snapshot_observed_at":"2026-08-17T20:49:11.320052Z","submitted_at":"2021-04-01T15:19:08Z","title":"Replay in Deep Learning: Current Approaches and Missing Biological Elements","version":2},"cited_work":{"arxiv_id":"2104.04132","doi":null,"metadata_source":"pith","pith_arxiv_id":"2104.04132","snapshot_observed_at":"2026-08-15T20:36:40.620699Z","title":"Replay in Deep Learning: Current Approaches and Missing Biological Elements","venue":"q-bio.NC","work_id":"45653c13-f03a-42cc-a9d9-7b64c86301c0","year":2021},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.741283Z"},"links":{"cited_paper":"/paper/2104.04132","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:dbc526f7e75ca43e5c3f4a037c094ff23a13d9001e75b2dab5af63466097f11a","observation_id":"495ee9ba-6910-41ed-afb2-c690a904f61f","resolution":{"observed_at":"2026-08-15T20:36:40.627276Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2404.10758","last_updated":"2024-05-09T21:03:58Z","snapshot_observed_at":"2026-08-16T14:00:23.290288Z","submitted_at":"2024-04-16T17:35:35Z","title":"Watch Your Step: Optimal Retrieval for Continual Learning at Scale","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.10758","snapshot_observed_at":"2026-08-15T20:36:39.748015Z","title":"Watch your step: Optimal retrieval for continual learning at scale.arXiv preprint arXiv:2404.10758, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.748015Z"},"links":{"cited_paper":"/paper/2404.10758","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:f4a3d9cc651577d400a0191322b11afe803d665dcc9272fbeeddec39ad9fc11f","observation_id":"97881e55-a423-42f2-9ce6-24de3d6228ea","resolution":{"observed_at":"2026-08-15T20:36:39.748015Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1503.02531","last_updated":"2015-03-09T15:44:49Z","snapshot_observed_at":"2026-08-16T18:00:58.008096Z","submitted_at":"2015-03-09T15:44:49Z","title":"Distilling the Knowledge in a Neural Network","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1503.02531","snapshot_observed_at":"2026-08-15T20:36:39.757280Z","title":"Distilling the knowledge in a neural network.ArXiv, abs/1503.02531, 2015","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.757280Z"},"links":{"cited_paper":"/paper/1503.02531","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:aa67b3aff8e973fe5efe1c9d7e0a0b6399ce76f0471e6a4a4fd43b236e92c202","observation_id":"33d4d77b-8cb5-4845-aa0f-ac8351a737b3","resolution":{"observed_at":"2026-08-15T20:36:39.757280Z","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":"2026-08-15T20:36:41.093879Z","title":"Parameter-efficient transfer learning for nlp","venue":null,"work_id":"a66460bf-351c-4119-94e6-84ee7c416219","year":2019},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.764504Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:5e64e0a32cdab257ab096fa09dbd5a3e07992b8f0440dc72387c4bf2c215efdf","observation_id":"c24566ed-89be-4730-9c1f-824c4bf8f4ec","resolution":{"observed_at":"2026-08-15T20:36:41.098287Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:41.080190Z","title":"Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen","venue":null,"work_id":"4f309ef6-fb21-44a4-ba06-239bf856c4d6","year":2021},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.770037Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:34a1cc913c3697baab421d0a7c9ea8d64b6df020c8cfab6f02a9ea9b20eadfe4","observation_id":"3224e36d-6780-43dc-9465-543921014df8","resolution":{"observed_at":"2026-08-15T20:36:41.084908Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2404.10981","last_updated":"2024-08-23T00:17:02Z","snapshot_observed_at":"2026-08-14T10:51:21.055151Z","submitted_at":"2024-04-17T01:27:42Z","title":"A Survey on Retrieval-Augmented Text Generation for Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.10981","snapshot_observed_at":"2026-08-15T20:36:39.774836Z","title":"A survey on retrieval- augmented text generation for large language models.arXiv preprint arXiv:2404.10981, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.774836Z"},"links":{"cited_paper":"/paper/2404.10981","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:3955f544213fed0fc79985f466916cbcacb9fa46b45461e5cb6c0723be55a4cc","observation_id":"6a5429c5-cdd9-4625-8a41-beb6d3d51615","resolution":{"observed_at":"2026-08-15T20:36:39.774836Z","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":"2026-08-15T20:36:41.065569Z","title":"Position: Open-endedness is essential for artificial superhuman intelligence","venue":null,"work_id":"c09dc7db-0aae-4cdd-8f07-7ce443aee026","year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.779833Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:a16f63c942cafbf858d370422bbc7a56ccf9f7d922466e4fc4a0cd9c6d94ae98","observation_id":"8888d7a0-a06f-4b9f-affa-669abc410381","resolution":{"observed_at":"2026-08-15T20:36:41.070832Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.08763","last_updated":"2024-09-04T16:13:18Z","snapshot_observed_at":"2026-08-16T14:10:04.613423Z","submitted_at":"2024-03-13T17:58:57Z","title":"Simple and Scalable Strategies to Continually Pre-train Large Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.08763","snapshot_observed_at":"2026-08-15T20:36:39.784001Z","title":"Richter, Quentin Anthony, Timoth ´ee Lesort, Eugene Belilovsky, and Irina Rish","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.784001Z"},"links":{"cited_paper":"/paper/2403.08763","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:7bc499e05bf27655761dd21eaa206763e8ff9e733d8898def8908bbe38fed47f","observation_id":"2c718bbd-17db-4aa6-b20e-c781404f3d96","resolution":{"observed_at":"2026-08-15T20:36:39.784001Z","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":"2026-08-15T20:36:41.049199Z","title":"Open- clip, 2021","venue":null,"work_id":"7ec253d0-6bfa-45fd-8c24-0e80e0baef6b","year":2021},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.788881Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:fd5205106381317095d908f314a0143d831c49f6eba043a9768bab2206ba3b9a","observation_id":"7e1a2f26-a80c-47e1-bb98-8354b09bd90d","resolution":{"observed_at":"2026-08-15T20:36:41.054406Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2212.04089","last_updated":"2023-03-31T15:27:01Z","snapshot_observed_at":"2026-08-13T03:27:01.609831Z","submitted_at":"2022-12-08T05:50:53Z","title":"Editing Models with Task Arithmetic","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.04089","snapshot_observed_at":"2026-08-15T20:36:39.793386Z","title":"Editing models with task arithmetic.arXiv preprint arXiv:2212.04089, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.793386Z"},"links":{"cited_paper":"/paper/2212.04089","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:5ec371324a3099964bb5bd1aa16de943355c24b46ac41159d84e8017f7f36ea7","observation_id":"e6d95999-68ee-400b-8c66-2b9bc6e87e53","resolution":{"observed_at":"2026-08-15T20:36:39.793386Z","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":"2026-08-15T20:36:41.036244Z","title":"Unlocking the power of function vectors for characterizing and mitigating catastrophic forgetting in continual instruction tuning","venue":null,"work_id":"b17b4713-c83b-4801-8d19-d272395aff4a","year":null},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.798482Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:c8fc0bee31ae536aee3c86c9f8f118f89cb70f996c996ef956208025965c70d5","observation_id":"61bac005-b78d-486d-9202-919e9cb24b05","resolution":{"observed_at":"2026-08-15T20:36:41.040915Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:41.020911Z","title":"Continual pre-training of lan- guage models","venue":null,"work_id":"8d2201ee-b5bb-4f8f-8b5e-6b0056772a44","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.803852Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:cf550f4919c1b920d9f6edcbb1f5c496d5515bef52696e4e4d2990fabd50317c","observation_id":"7f555cfc-9042-4487-a01a-3751b809df4a","resolution":{"observed_at":"2026-08-15T20:36:41.025125Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:41.004571Z","title":"Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A","venue":null,"work_id":"8e2011ee-a7ca-439e-93c9-5220fc203d8a","year":2017},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.808375Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:1441f0410aab8e6a3a82f9f9ad558e30082a1a39e289a2c205b3ff68325899d9","observation_id":"f0dd6fb5-fd24-45e8-af91-82822e4cdef4","resolution":{"observed_at":"2026-08-15T20:36:41.010130Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.991668Z","title":"McClel- land","venue":null,"work_id":"8fa9a458-8ffc-460a-ad84-ed54776545fc","year":2016},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.812304Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:fae13bf2c6e9b53c4c4afe3de3789c6c7628168fe4f6053fe2dfc65ea5875888","observation_id":"13765f42-efc1-43d7-a60e-41d34c6af33d","resolution":{"observed_at":"2026-08-15T20:36:40.995704Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.976298Z","title":"The power of scale for parameter-efficient prompt tuning","venue":null,"work_id":"c65d7c9d-55b4-4d81-8ae0-7de67ee2079e","year":2021},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.817843Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:eb3902eb3d2580d48547442a5bb2b0b6206ccb9bd211ad6c5fe18f61e04c8653","observation_id":"1d6be6db-a26c-482c-bf29-50f6b9ca3087","resolution":{"observed_at":"2026-08-15T20:36:40.980529Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2411.01775","last_updated":"2024-11-04T03:54:00Z","snapshot_observed_at":"2026-08-16T13:03:20.121177Z","submitted_at":"2024-11-04T03:54:00Z","title":"Eurekaverse: Environment Curriculum Generation via Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2411.01775","snapshot_observed_at":"2026-08-15T20:36:39.822334Z","title":"Eurekaverse: Environment curriculum generation via large language mod- els.arXiv preprint arXiv:2411.01775, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.822334Z"},"links":{"cited_paper":"/paper/2411.01775","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:63472921ed774c5c68555c05207830b8630cbfe97a2ef7878b2005006ab95a1f","observation_id":"3d4106fb-e760-4675-9a09-5a98e81deb17","resolution":{"observed_at":"2026-08-15T20:36:39.822334Z","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":"2026-08-15T20:36:40.963414Z","title":"Loss decoupling for task- agnostic continual learning","venue":null,"work_id":"08fe2dea-df5c-4871-9969-75e346896dd5","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.827875Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:fd649d3294406c1804a98ebee2f6ceddf58cd3027d49837a803b81dc0ea77ddf","observation_id":"3827d59c-ad38-4f1a-9a14-cb402e741ab5","resolution":{"observed_at":"2026-08-15T20:36:40.967764Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2502.07978","last_updated":"2025-02-11T21:52:19Z","snapshot_observed_at":"2026-08-12T13:45:06.405379Z","submitted_at":"2025-02-11T21:52:19Z","title":"A Survey of In-Context Reinforcement Learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.07978","snapshot_observed_at":"2026-08-15T20:36:39.832090Z","title":"A survey of in-context reinforcement learning.arXiv preprint arXiv:2502.07978, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.832090Z"},"links":{"cited_paper":"/paper/2502.07978","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:04d4a53b8a602cf91617fda8f5f701c6c1ea62987c04c787bd78042179a8a741","observation_id":"c1f271fa-c67b-46e2-bd0a-1e4a7ffcc0b4","resolution":{"observed_at":"2026-08-15T20:36:39.832090Z","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":"2026-08-15T20:36:40.947737Z","title":"Task arithmetic in the tangent space: Improved editing of pre-trained models","venue":null,"work_id":"8443a36b-bbd2-4ebf-bb41-676d8f60434b","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.837751Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:8a061e27c77316d20774cdf3170ecda1a06e6bcd503436432062385ace5fb3b9","observation_id":"455113d9-aa82-437a-a1f9-e2718abeca46","resolution":{"observed_at":"2026-08-15T20:36:40.952431Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.12957","last_updated":"2025-04-03T10:12:23Z","snapshot_observed_at":"2026-08-16T13:33:18.623066Z","submitted_at":"2024-07-17T18:59:56Z","title":"R+X: Retrieval and Execution from Everyday Human Videos","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.12957","snapshot_observed_at":"2026-08-15T20:36:39.842118Z","title":"R+x: Retrieval and execution from everyday human videos.arXiv preprint arXiv:2407.12957, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.842118Z"},"links":{"cited_paper":"/paper/2407.12957","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:0f8a266c83b72a8a20173190666ad8e76cf86a9439c867ea88361da308552624","observation_id":"28fd64b7-a0d6-4a7f-9e56-ffd91e4277df","resolution":{"observed_at":"2026-08-15T20:36:39.842118Z","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":"2026-08-15T20:36:40.931162Z","title":"Berg, and Li Fei-Fei","venue":null,"work_id":"141b1302-556c-49f7-9c33-d0b754805463","year":2015},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.850282Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:c8e2f89a93e769749c8b71a42d175d64e659d94d60d60be8b3506d40cdb17972","observation_id":"98194bbd-0b94-46e3-9f1f-33ed707f5209","resolution":{"observed_at":"2026-08-15T20:36:40.936323Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1606.04671","last_updated":"2022-10-22T14:34:44Z","snapshot_observed_at":"2026-08-16T23:40:37.342111Z","submitted_at":"2016-06-15T08:20:51Z","title":"Progressive Neural Networks","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1606.04671","snapshot_observed_at":"2026-08-15T20:36:39.855457Z","title":"Rusu, Neil C","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.855457Z"},"links":{"cited_paper":"/paper/1606.04671","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:223f3a3c5a15b75ccd3ede3653ccb7a2f2add238a1e0db300f5b8f73c6f08573","observation_id":"0c0baaa2-7591-4ba1-b17a-d09cb7dbf5ca","resolution":{"observed_at":"2026-08-15T20:36:39.855457Z","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":"2026-08-15T20:36:40.916634Z","title":null,"venue":null,"work_id":"c2f5cb9e-6de1-410f-91a8-dee49412fbcf","year":2025},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.860939Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:4dec930419089ceb0c8a9e9507c38c45465d7c4d274580f3f6fd18acb2fe95f0","observation_id":"c7cd136f-cd94-4286-ac17-1aa84e3c8e58","resolution":{"observed_at":"2026-08-15T20:36:40.921579Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2203.17269","last_updated":"2023-04-03T22:49:29Z","snapshot_observed_at":"2026-08-16T17:10:09.872612Z","submitted_at":"2022-03-31T17:59:00Z","title":"A Closer Look at Rehearsal-Free Continual Learning","version":2},"cited_work":{"arxiv_id":"2203.17269","doi":null,"metadata_source":"pith","pith_arxiv_id":"2203.17269","snapshot_observed_at":"2026-08-15T20:36:40.394540Z","title":"A Closer Look at Rehearsal-Free Continual Learning","venue":"cs.LG","work_id":"94a7f4d5-3dfe-4e2f-89bd-b90fa197da54","year":2022},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.865942Z"},"links":{"cited_paper":"/paper/2203.17269","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:6a41fabbde84ee9808c345c2fa9ded8137a5380e3560ea80fef1d93aa9e38aea","observation_id":"15020da6-b3f2-429a-81ca-394ab6548143","resolution":{"observed_at":"2026-08-15T20:36:40.402912Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.03314","last_updated":"2024-08-06T17:35:05Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-08-06T17:35:05Z","title":"Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.03314","snapshot_observed_at":"2026-08-15T20:36:39.872490Z","title":"Scaling llm test-time compute optimally can be more effective than scaling model parameters.arXiv preprint arXiv:2408.03314, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.872490Z"},"links":{"cited_paper":"/paper/2408.03314","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:09e25443e97f842cde8b8943411f0a317bf8d0855579b55ac820ce39c4db916d","observation_id":"e05a5980-71c8-4beb-a424-7a969622d53a","resolution":{"observed_at":"2026-08-15T20:36:39.872490Z","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":"2026-08-15T20:36:40.900542Z","title":"Improving online continual learning performance and stability with temporal ensembles","venue":null,"work_id":"dda175b1-a56b-486c-84cf-f6a08e693f84","year":null},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.877626Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:082ab29b29df1e0470432cc9eff8c59c7b8860ca34398659eaf9f78e94d2a2f7","observation_id":"4dd46f1b-11be-4283-b32a-6f773cc1f83d","resolution":{"observed_at":"2026-08-15T20:36:40.905420Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.886311Z","title":"Logit standardization in knowledge distillation","venue":null,"work_id":"15a46ffb-ced1-4a40-9522-d46a7812f984","year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.883576Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:e26ee9b1abd9dbaaa7557cd02cd66507043eff5d826cbf6f37052f4d5608feae","observation_id":"65709ad1-695d-4403-ace5-438c5a833753","resolution":{"observed_at":"2026-08-15T20:36:40.891410Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1904.07734","last_updated":"2019-04-15T12:22:36Z","snapshot_observed_at":"2026-08-14T16:46:31.118689Z","submitted_at":"2019-04-15T12:22:36Z","title":"Three scenarios for continual learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1904.07734","snapshot_observed_at":"2026-08-15T20:36:39.888750Z","title":"Three scenar- ios for continual learning.arXiv preprint arXiv:1904.07734, 2019","venue":null,"work_id":null,"year":1904},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.888750Z"},"links":{"cited_paper":"/paper/1904.07734","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:249ebe860d4f6424b97badfec93d893e88109fc0f6fdb7721d2d7cd732f16f63","observation_id":"b9a7c7a7-16bb-434c-9c01-90c1c7832ad5","resolution":{"observed_at":"2026-08-15T20:36:39.888750Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2311.11908","last_updated":"2024-03-28T13:16:50Z","snapshot_observed_at":"2026-08-16T14:41:47.500694Z","submitted_at":"2023-11-20T16:40:29Z","title":"Continual Learning: Applications and the Road Forward","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2311.11908","snapshot_observed_at":"2026-08-15T20:36:39.895084Z","title":"Hayes, Eyke H ¨ullermeier, Christopher Kanan, Dhireesha Kudithipudi, Christoph H","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.895084Z"},"links":{"cited_paper":"/paper/2311.11908","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:b20574d709748f62842c865f9f5f2dbac3ceb5a0be913413e16d77486d38d247","observation_id":"39782e8a-3ebb-47d3-908d-e82e2a40511a","resolution":{"observed_at":"2026-08-15T20:36:39.895084Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2311.02058","last_updated":"2024-11-23T06:28:06Z","snapshot_observed_at":"2026-08-17T20:51:39.805864Z","submitted_at":"2023-11-03T17:38:35Z","title":"LOTUS: Continual Imitation Learning for Robot Manipulation Through Unsupervised Skill Discovery","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2311.02058","snapshot_observed_at":"2026-08-15T20:36:39.902351Z","title":"Lotus: Continual imitation learning for robot manipula- tion through unsupervised skill discovery.arXiv preprint arXiv:2311.02058, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.902351Z"},"links":{"cited_paper":"/paper/2311.02058","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:de2b1c08f86d115641aaf77bd2b16bdf858c8c6863da894402b1957f0f8f08f9","observation_id":"f1aaca85-2405-4aba-b369-c9fb4944ccae","resolution":{"observed_at":"2026-08-15T20:36:39.902351Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.16291","last_updated":"2023-10-19T16:27:03Z","snapshot_observed_at":"2026-08-13T16:55:46.498156Z","submitted_at":"2023-05-25T17:46:38Z","title":"Voyager: An Open-Ended Embodied Agent with Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.16291","snapshot_observed_at":"2026-08-15T20:36:39.908612Z","title":"V oyager: An open-ended embodied agent with large language models.arXiv preprint arXiv:2305.16291, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.908612Z"},"links":{"cited_paper":"/paper/2305.16291","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:92a139805482b4eabd9475896b3f9ac401a2ce6490147204cd717ff89f8afcdb","observation_id":"1b7527e8-1669-4f8f-8b44-a4d7c65436ea","resolution":{"observed_at":"2026-08-15T20:36:39.908612Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.00487","last_updated":"2024-02-06T09:12:09Z","snapshot_observed_at":"2026-08-16T15:58:52.866918Z","submitted_at":"2023-01-31T11:34:56Z","title":"A Comprehensive Survey of Continual Learning: Theory, Method and Application","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.00487","snapshot_observed_at":"2026-08-15T20:36:39.913971Z","title":"A comprehensive survey of continual learning: Theory, method and application.arXiv preprint arXiv:2302.00487, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.913971Z"},"links":{"cited_paper":"/paper/2302.00487","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:ef82457f0556bfdcb63d7dfbbeb933a709f527a15ea0cbe3a977b9ea6b871504","observation_id":"245eb1cf-096e-44af-b060-9da2b18229c3","resolution":{"observed_at":"2026-08-15T20:36:39.913971Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.05229","last_updated":"2025-04-16T23:04:42Z","snapshot_observed_at":"2026-08-16T13:36:29.702453Z","submitted_at":"2024-07-07T01:50:25Z","title":"HiDe-PET: Continual Learning via Hierarchical Decomposition of Parameter-Efficient Tuning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.05229","snapshot_observed_at":"2026-08-15T20:36:39.919886Z","title":"Hide-pet: Continual learning via hierarchical de- composition of parameter-efficient tuning.arXiv preprint arXiv:2407.05229, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.919886Z"},"links":{"cited_paper":"/paper/2407.05229","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:c763461b9d20e6a1c526481fcc393c3a26ba6634689a35ff3c03706c25a88790","observation_id":"c280793e-2621-45d2-949f-31769d41da21","resolution":{"observed_at":"2026-08-15T20:36:39.919886Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.07617","last_updated":"2026-05-31T18:45:03Z","snapshot_observed_at":"2026-08-17T13:43:29.791103Z","submitted_at":"2025-02-11T15:05:33Z","title":"Scaling Pre-training to One Hundred Billion Data for Vision Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.07617","snapshot_observed_at":"2026-08-15T20:36:39.924686Z","title":"Scaling pre-training to one hundred billion data for vision language models.arXiv preprint arXiv:2502.07617, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.924686Z"},"links":{"cited_paper":"/paper/2502.07617","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:00fbd52c6a0b8fc39c2cf8f3f4766ecfa76829e51001be21480a7b250f402ef5","observation_id":"cad0bd5e-de43-4592-b323-ef15ef4e9faf","resolution":{"observed_at":"2026-08-15T20:36:39.924686Z","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":"2026-08-15T20:36:40.871496Z","title":"Dualprompt: Com- plementary prompting for rehearsal-free continual learning","venue":null,"work_id":"d0c375e8-78ca-43d7-aba8-5e2137e1cf82","year":2022},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.929448Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:a8c56686e39081171849e8584ff7bc795e6d91ad28d6496476fdbe8f5f711a52","observation_id":"2a030f35-316d-49a6-b1a1-a327d3a076a4","resolution":{"observed_at":"2026-08-15T20:36:40.876702Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.857771Z","title":"Learning to prompt for con- tinual learning","venue":null,"work_id":"914372f8-7b0e-4048-818a-bad50cae4dd6","year":2022},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.933407Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:9420a2316888f0c1400459c527bdc7a307403e5c166b0ef6d931daba08f30987","observation_id":"747174de-b0c3-410a-81e1-7a763d9dbf49","resolution":{"observed_at":"2026-08-15T20:36:40.862286Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2311.17601","last_updated":"2023-11-29T12:53:32Z","snapshot_observed_at":"2026-08-17T17:50:09.159949Z","submitted_at":"2023-11-29T12:53:32Z","title":"Continual Learning with Low Rank Adaptation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2311.17601","snapshot_observed_at":"2026-08-15T20:36:39.938877Z","title":"Continual learning with low rank adap- tation.arXiv preprint arXiv:2311.17601, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.938877Z"},"links":{"cited_paper":"/paper/2311.17601","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:0a915648c252379e74c3898a390f75ac484c477187521e4c928b595ad683756d","observation_id":"2701cef0-8d5b-4a49-aaa9-7a540a88dae8","resolution":{"observed_at":"2026-08-15T20:36:39.938877Z","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":"2026-08-15T20:36:40.842364Z","title":null,"venue":null,"work_id":"01f784b1-ef41-40e6-9097-343ed8155fe1","year":2020},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.943806Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:37a4cb10d5377801ebcb38a7fb54ff40b8d2077bf22a779ab431877e676f0c0b","observation_id":"75944516-8a84-42d9-ab38-082a0b82108a","resolution":{"observed_at":"2026-08-15T20:36:40.847535Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.828066Z","title":"Ties-merging: Resolving interference when merging models","venue":null,"work_id":"6acf2fa2-cff3-4033-b8df-4a89c9e2d88c","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.948484Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:c32fcf4879b16e6ff759b6ef06f582f1c9a1fb6f84e89531002bc1b05edab6b7","observation_id":"a95ecb7e-e125-47fd-bedd-b7e95b2b3394","resolution":{"observed_at":"2026-08-15T20:36:40.832841Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.03617","last_updated":"2024-10-04T17:17:19Z","snapshot_observed_at":"2026-08-16T13:12:30.478509Z","submitted_at":"2024-10-04T17:17:19Z","title":"What Matters for Model Merging at Scale?","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.03617","snapshot_observed_at":"2026-08-15T20:36:39.952965Z","title":"What matters for model merging at scale?arXiv preprint arXiv:2410.03617, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.952965Z"},"links":{"cited_paper":"/paper/2410.03617","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:8f8c402f3bf20cee1dbd42fb5166e551cf693f4eba0699dfbc70dd0114be6001","observation_id":"09021c6b-cf5b-4ed8-9934-44ffd9b31f61","resolution":{"observed_at":"2026-08-15T20:36:39.952965Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2408.07666","last_updated":"2025-12-31T04:06:49Z","snapshot_observed_at":"2026-08-07T23:28:24.025478Z","submitted_at":"2024-08-14T16:58:48Z","title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.07666","snapshot_observed_at":"2026-08-15T20:36:39.958433Z","title":"Model merging in llms, mllms, and beyond: Methods, theories, applications and opportunities.arXiv preprint arXiv:2408.07666, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.958433Z"},"links":{"cited_paper":"/paper/2408.07666","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:24909f8ef260cd29859687a3603be4b372e91f6afc262f9b5595765898d63973","observation_id":"684d57f5-55f6-449c-9788-0f61bc491016","resolution":{"observed_at":"2026-08-15T20:36:39.958433Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2306.12026","last_updated":"2023-06-21T05:26:28Z","snapshot_observed_at":"2026-08-17T03:10:05.882159Z","submitted_at":"2023-06-21T05:26:28Z","title":"Continual Learners are Incremental Model Generalizers","version":1},"cited_work":{"arxiv_id":"2306.12026","doi":null,"metadata_source":"pith","pith_arxiv_id":"2306.12026","snapshot_observed_at":"2026-08-15T20:36:40.140739Z","title":"Continual Learners are Incremental Model Generalizers","venue":"cs.LG","work_id":"91808962-f294-4d11-aac9-fb2bde876322","year":2023},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.963383Z"},"links":{"cited_paper":"/paper/2306.12026","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:8ec997309c92ce143ef6a5ce18a07b2a31e62f4f5202079397a6c8f828d53e05","observation_id":"5a7b8dc7-2c9d-47b7-af29-680c90fe9ff4","resolution":{"observed_at":"2026-08-15T20:36:40.147891Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.809257Z","title":"Boosting continual learning of vision-language models via mixture-of-experts adapters","venue":null,"work_id":"67938a6b-002c-41eb-b181-9c1c454f683b","year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.968019Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:0e7acee290d71135751f6b6f8e4a993c641088e947ad1ea61597b8dd5f04a2c3","observation_id":"b7fb3d7e-7836-4b2f-96f2-ca3acf4d7ecf","resolution":{"observed_at":"2026-08-15T20:36:40.815167Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.789345Z","title":"Language models are super mario: Absorbing abilities from homologous models as a free lunch","venue":null,"work_id":"2e9183ba-2080-4014-ab0a-bac82e18dd4c","year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.977613Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:1fbb5a4bd5918afe2aec0a5f2159f7f915c0505ebdb114117cec654bb48194cc","observation_id":"682bb8c4-0519-4a91-aed8-9bffb16f6c31","resolution":{"observed_at":"2026-08-15T20:36:40.798360Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.17193","last_updated":"2024-02-27T04:18:49Z","snapshot_observed_at":"2026-08-16T14:15:03.626390Z","submitted_at":"2024-02-27T04:18:49Z","title":"When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Method","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.17193","snapshot_observed_at":"2026-08-15T20:36:39.982811Z","title":"When scaling meets llm finetuning: The effect of data, model and finetuning method.arXiv preprint arXiv:2402.17193, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.982811Z"},"links":{"cited_paper":"/paper/2402.17193","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:3bb5f640d359dd65aa854ec7ac04b0898293b1333d3f53ae80eeb93f8e381005","observation_id":"49af8606-e438-4229-aee3-1729c705c914","resolution":{"observed_at":"2026-08-15T20:36:39.982811Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.17920","last_updated":"2025-02-25T07:35:36Z","snapshot_observed_at":"2026-08-16T12:55:28.609032Z","submitted_at":"2025-02-25T07:35:36Z","title":"C-LoRA: Continual Low-Rank Adaptation for Pre-trained Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.17920","snapshot_observed_at":"2026-08-15T20:36:39.988068Z","title":"C-lora: Continual low-rank adaptation for pre-trained models.arXiv preprint arXiv:2502.17920, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.988068Z"},"links":{"cited_paper":"/paper/2502.17920","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:ab21197366887a5f290b06dc45ace94bcc116cbc873e971530a516b05f384d5a","observation_id":"738483fd-c6de-459c-bfb7-471a867d4dcf","resolution":{"observed_at":"2026-08-15T20:36:39.988068Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.12048","last_updated":"2024-02-19T11:02:05Z","snapshot_observed_at":"2026-08-16T14:17:22.875988Z","submitted_at":"2024-02-19T11:02:05Z","title":"Model Tailor: Mitigating Catastrophic Forgetting in Multi-modal Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.12048","snapshot_observed_at":"2026-08-15T20:36:39.994817Z","title":"Model tailor: Mitigating catastrophic forgetting in multi-modal large lan- guage models.arXiv preprint arXiv:2402.12048, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:39.994817Z"},"links":{"cited_paper":"/paper/2402.12048","citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:b2d13933bdd5b891cdff98dbb4c69be2e391e45c2a92ed08fb03b4c6607f09f2","observation_id":"561bb109-b054-4d2f-8d3e-f75c5540c614","resolution":{"observed_at":"2026-08-15T20:36:39.994817Z","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":"2026-08-15T20:36:40.771097Z","title":null,"venue":null,"work_id":"3959aeeb-2cc2-49fb-8a02-850fabe38457","year":null},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:40.001203Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:2720ec9261a0cd2f0936c69270a2df306468c2f59da63a58fbac1899b432f6d4","observation_id":"3d221bd1-4184-4cdc-8350-88d2d9ec76a2","resolution":{"observed_at":"2026-08-15T20:36:40.776185Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.755719Z","title":null,"venue":null,"work_id":"bdae0288-5690-4fcb-9930-169cf0c61536","year":null},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:40.011692Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:c0a71c1bc16642f8229c44161dbeaa3140c9bc536448ab118f8bcadd81ccb4ce","observation_id":"8321aee1-8b30-488a-867c-8c0068008d87","resolution":{"observed_at":"2026-08-15T20:36:40.760425Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+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":"2026-08-15T20:36:40.741634Z","title":null,"venue":null,"work_id":"fa311837-d421-4584-8308-b1215e24c134","year":null},"citing_paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-15T20:36:40.017952Z"},"links":{"citing_paper":"/paper/2505.12512"},"observation_digest":"sha256:62eb8be281a88e8c35773a828cce3ac4ab714fbd62cd3cd5c930ec15b6258c6d","observation_id":"96c8ffc4-0274-4c01-a541-9b3875b9665a","resolution":{"observed_at":"2026-08-15T20:36:40.745781Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2505.12512","last_updated":"2025-05-18T18:23:50Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-17T20:38:39.522719Z","submitted_at":"2025-05-18T18:23:50Z","title":"Scalable Strategies for Continual Learning with Replay"},"reference_resolution":{"displayed":61,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":35,"verified_exact":3,"verified_fuzzy":23},"total_outbound_references":61},"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":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 61 of 61 outbound references and 2 inbound Pith citation observations for arXiv:2505.12512."}