{"as_of":"2026-08-08T09:08:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:1b8d20679976237a698ed1bc2452bbbd9c2f361ca909bd481522cca906383524","coverage":[{"denominator":38,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":38,"source":"paper_references, paper_reference_links","source_observed_at":"2026-07-31T12:39:14.368486Z","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":"2026-08-08T06:32:00.761636+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T00:14:14.565512Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"pith","source_observed_at":"2026-08-07T00:14:19.529738Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"cited_work":{"arxiv_id":"2607.28272","doi":null,"metadata_source":"pith","pith_arxiv_id":"2607.28272","snapshot_observed_at":"2026-08-07T00:14:19.529738Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","venue":"cs.AI","work_id":"a4a670de-318c-4c60-a353-7c6c75de5c88","year":2026},"citing_paper":{"arxiv_id":"2608.02713","last_updated":"2026-08-03T17:59:58Z","snapshot_observed_at":"2026-08-07T23:09:56.055766Z","submitted_at":"2026-08-03T17:59:58Z","title":"Quo Vadis, World Modeling?","version":1},"reference_index":174,"source":"pdf_text","source_observed_at":"2026-08-07T00:14:14.565512Z"},"links":{"cited_paper":"/paper/2607.28272","citing_paper":"/paper/2608.02713"},"observation_digest":"sha256:a5e790989f7b0fb9f56e4ec09981dc863529fede754f83181d86f1acac771188","observation_id":"15317e70-87c0-48d3-bbb8-4649e4134d12","resolution":{"observed_at":"2026-08-07T00:14:19.609655Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2607.28272/citation-record","integrity":"/paper/2607.28272/integrity","json":"/paper/2607.28272/citation-record.json","paper":"/paper/2607.28272"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2305.11598","last_updated":"2023-05-19T11:20:37Z","snapshot_observed_at":"2026-08-05T10:48:49.527192Z","submitted_at":"2023-05-19T11:20:37Z","title":"Introspective Tips: Large Language Model for In-Context Decision Making","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.11598","snapshot_observed_at":"2026-07-31T12:39:13.979151Z","title":"Introspective tips: Large language model for in-context decision making.arXiv preprint arXiv:2305.11598, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:13.979151Z"},"links":{"cited_paper":"/paper/2305.11598","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:34f9ca9cfe1b2124021a6bbeb3a175d05c546c3b1dcc0183ff7dbce65a4724d1","observation_id":"a18c6404-ad7b-4053-94c1-239f135d712e","resolution":{"observed_at":"2026-07-31T12:39:13.979151Z","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":"2026-07-31T12:39:13.993771Z","title":"Re-searcher: Robust agentic search with goal-oriented planning and self-reflection","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:13.993771Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:fe08921edb94e2153a80b075bc1bd4571e74c81e7f99db7ac5566a4d6b237cea","observation_id":"de88d169-e33f-42db-84fb-bd6104d45256","resolution":{"observed_at":"2026-07-31T12:39:13.993771Z","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":"2026-07-31T12:39:14.005622Z","title":"Search-r1: Training llms to reason and leverage search engines with reinforcement learning, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.005622Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:7c9b7c245edf71e46734a3ce7e854ac91a0b139e725642b31e33316bc333b9af","observation_id":"4036f96b-0a3a-4922-bc27-bc058bed6d9e","resolution":{"observed_at":"2026-07-31T12:39:14.005622Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2505.16582","last_updated":"2025-05-26T10:07:05Z","snapshot_observed_at":"2026-08-07T14:56:04.605737Z","submitted_at":"2025-05-22T12:17:13Z","title":"O$^2$-Searcher: A Searching-based Agent Model for Open-Domain Open-Ended Question Answering","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.16582","snapshot_observed_at":"2026-07-31T12:39:14.014060Z","title":"o2-searcher: A searching-based agent model for open-domain open-ended question answering.arXiv preprint arXiv:2505.16582, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.014060Z"},"links":{"cited_paper":"/paper/2505.16582","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:1af3050c5a0c759e78d3c4e7ce90b6c4371776cc9b6137ffaf24397b9558f4bc","observation_id":"09a51526-e474-4168-9506-abd0a408f3fd","resolution":{"observed_at":"2026-07-31T12:39:14.014060Z","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":"2026-07-31T12:39:14.022617Z","title":"A comprehensive survey of continual learning: Theory, method and application, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.022617Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:0f00b64480e58ede57f48aaa2d8fbc77074e329ba6470c68da20a3b46baf8b6f","observation_id":"47e63ae8-7773-4860-b15d-3abcca0b5008","resolution":{"observed_at":"2026-07-31T12:39:14.022617Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19828","last_updated":"2026-01-14T14:21:21Z","snapshot_observed_at":"2026-07-06T22:19:28.487854Z","submitted_at":"2025-08-27T12:26:55Z","title":"Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.19828","snapshot_observed_at":"2026-07-31T12:39:14.029343Z","title":"Memory-r1: Enhancing large language model agents to manage and utilize memories via reinforcement learning.arXiv preprint arXiv:2508.19828, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.029343Z"},"links":{"cited_paper":"/paper/2508.19828","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:43688179de45910e0af504335a2929ee3fe8b6cd83c7075a827e835d907b13f0","observation_id":"fe573b4c-663f-4fac-8b7d-838deee0b0cc","resolution":{"observed_at":"2026-07-31T12:39:14.029343Z","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":"2026-07-31T12:39:14.042424Z","title":"A-mem: Agentic memory for llm agents.Advances in Neural Information Processing Systems, 38:17577–17604, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.042424Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:59e7512e4bf07c7801ff03c9d5fa81031b3c11b1c7b5f7d01226b080b0e30dd9","observation_id":"c26a3c47-e8fc-4cce-b928-511d175f2a59","resolution":{"observed_at":"2026-07-31T12:39:14.042424Z","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":"2026-07-31T12:39:14.051183Z","title":"Towards self-evolving agents: Enabling autonomy through interactive experience refinement","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.051183Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:7aa1d3b38a2393ac1dc731cab1626f63d36f5d5c49fa5ae97166b73938d11760","observation_id":"5b6e151f-6dd7-40b1-9909-f98bc1c9763b","resolution":{"observed_at":"2026-07-31T12:39:14.051183Z","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":"2026-07-31T12:39:14.058265Z","title":"G-memory: Tracing hierarchical memory for multi-agent systems, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.058265Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:11aaca175808cc0497f7060c38dc58e9dbd1890d047d25191ffb22b027dff2a8","observation_id":"d945a192-b391-46db-82c5-4aa68951cba1","resolution":{"observed_at":"2026-07-31T12:39:14.058265Z","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":"2026-07-31T12:39:14.068402Z","title":"Memgen: Weaving generative latent memory for self-evolving agents.arXiv preprint arXiv:2509.24704, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.068402Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:bb8e8943638c422d1449e11ff6ac757265cd27ef971ab351bbffb981ec8afe2b","observation_id":"50481e2d-f1d2-4042-96ab-095a1b4c568e","resolution":{"observed_at":"2026-07-31T12:39:14.068402Z","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":"2026-07-31T12:39:14.085217Z","title":"Reconstruction of automobile destruction: An example of the interaction between language and memory.Journal of verbal learning and verbal behavior, 13(5):585–589, 1974","venue":null,"work_id":null,"year":1974},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.085217Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:966e7fe3da2aeae068a126a23f9982ad9539e794d1a7ac17a15d14e111ab8b81","observation_id":"7c850c88-fb5e-449a-ba99-326bcc1de44a","resolution":{"observed_at":"2026-07-31T12:39:14.085217Z","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":"2026-07-31T12:39:14.091507Z","title":"React: Synergizing reasoning and acting in language models, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.091507Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:31d762160f40f8d1623e1482d6be2e5b0b9e01eb747edb7e0fffab12f2f7262f","observation_id":"cf80551f-dbf5-427d-ae85-934840faff82","resolution":{"observed_at":"2026-07-31T12:39:14.091507Z","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":"2026-07-31T12:39:14.101535Z","title":"Chain-of-thought prompting elicits reasoning in large language models.Advances in neural information processing systems, 35:24824–24837, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.101535Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:5770398d5f938db308ade0c4fc0f3bfe33d3d62f29e1f1f3b0cdf6db264a79a7","observation_id":"915a783e-a950-4768-8e71-743cf8a7f921","resolution":{"observed_at":"2026-07-31T12:39:14.101535Z","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":"2026-07-31T12:39:14.112117Z","title":"Reflexion: Language agents with verbal reinforcement learning, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.112117Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:912487d706a6b54b28959ab6a81c582f423d49ad21a40fed323453f43558df48","observation_id":"ccf978cb-135a-44ff-bcf0-9cafca7be93a","resolution":{"observed_at":"2026-07-31T12:39:14.112117Z","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":"2026-07-31T12:39:14.121965Z","title":"Correctad: A self-correcting agentic system to improve end-to-end planning in autonomous driving","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.121965Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:20cdd353ca95313ddda22ac01687b096a1f6cd7f26dfe0f7d1849649b4504662","observation_id":"b3eddf87-eab4-4d92-ae99-865fd66be7ad","resolution":{"observed_at":"2026-07-31T12:39:14.121965Z","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":"2026-07-31T12:39:14.142743Z","title":"Agent-flan: Designing data and methods of effective agent tuning for large language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.142743Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:1335a9d090ddb65955ccf52172f55eda201cd4f961fd2ae346a25356b79ffba4","observation_id":"0fa38185-fa82-4721-912c-4dbbb45d8b91","resolution":{"observed_at":"2026-07-31T12:39:14.142743Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.05915","last_updated":"2023-10-09T17:58:38Z","snapshot_observed_at":"2026-08-05T18:32:49.850038Z","submitted_at":"2023-10-09T17:58:38Z","title":"FireAct: Toward Language Agent Fine-tuning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.05915","snapshot_observed_at":"2026-07-31T12:39:14.155269Z","title":"Fireact: Toward language agent fine-tuning.arXiv preprint arXiv:2310.05915, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.155269Z"},"links":{"cited_paper":"/paper/2310.05915","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:51233f1b710678f1ec7031812fc668655668d364bce1b698a49f5373bcbc7595","observation_id":"6f20d570-efd7-4c93-ab90-7707693ad3a7","resolution":{"observed_at":"2026-07-31T12:39:14.155269Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1707.06347","last_updated":"2017-08-28T09:20:06Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-07-20T02:32:33Z","title":"Proximal Policy Optimization Algorithms","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1707.06347","snapshot_observed_at":"2026-07-31T12:39:14.161655Z","title":"Proximal policy optimization algorithms.arXiv preprint arXiv:1707.06347, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.161655Z"},"links":{"cited_paper":"/paper/1707.06347","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:63888543f4230d494fbbc1006af8485fb3c521a5221bfb717bf1c11a4bc2ca71","observation_id":"1e2233f0-20a4-4849-97c2-a7268a72e52b","resolution":{"observed_at":"2026-07-31T12:39:14.161655Z","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":"2026-07-31T12:39:14.168144Z","title":"Back to basics: Revisiting REINFORCE-style optimization for learning from human feedback in LLMs","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.168144Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:e9a156d945f83e6561b9d3f8d059fbdd98f9917095494e67ac652c6072f5219b","observation_id":"58f6ff02-ac4c-4ad9-9ae5-551f3363353d","resolution":{"observed_at":"2026-07-31T12:39:14.168144Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.03300","last_updated":"2024-04-27T15:25:53Z","snapshot_observed_at":"2026-08-06T14:58:42.911363Z","submitted_at":"2024-02-05T18:55:32Z","title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.03300","snapshot_observed_at":"2026-07-31T12:39:14.184750Z","title":"Deepseekmath: Pushing the limits of mathematical reasoning in open language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.184750Z"},"links":{"cited_paper":"/paper/2402.03300","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:954878d5af950b19b4c66602ad08f3c54aac73571699d63cf5d7768de4630004","observation_id":"fbabd18a-84db-4fb0-9269-d03923f21f69","resolution":{"observed_at":"2026-07-31T12:39:14.184750Z","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":"2026-07-31T12:39:14.194039Z","title":"Group-in-group policy optimization for llm agent training.Advances in Neural Information Processing Systems, 38:46375–46408, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.194039Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:27ea0e9372e11835c6b1c629989f57743ac664ea0f061dacdfefafcf4ae8165d","observation_id":"937c00e7-5500-4439-a158-d7f8c0ce1f18","resolution":{"observed_at":"2026-07-31T12:39:14.194039Z","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":"2026-07-31T12:39:14.209402Z","title":"Expel: Llm agents are experiential learners, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.209402Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:c28c55b94f446c3c90878d95dc14d8a1293cd75db7a8769562d15efef2b2a0f8","observation_id":"49f68abb-8aa5-44fe-93fd-83a8b3803a3c","resolution":{"observed_at":"2026-07-31T12:39:14.209402Z","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":"2026-07-31T12:39:14.218469Z","title":"Memp: Exploring agent procedural memory","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.218469Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:b83ad5acc5f077489d2e38fc1adee17c940f81cf7b6d8b95bd96aefe47b0a0fe","observation_id":"610905a6-994c-4365-9c19-ab3389f42972","resolution":{"observed_at":"2026-07-31T12:39:14.218469Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2504.19413","last_updated":"2025-04-28T01:46:35Z","snapshot_observed_at":"2026-08-02T07:32:11.339534Z","submitted_at":"2025-04-28T01:46:35Z","title":"Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2504.19413","snapshot_observed_at":"2026-07-31T12:39:14.229527Z","title":"Mem0: Building production- ready ai agents with scalable long-term memory.arXiv preprint arXiv:2504.19413, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.229527Z"},"links":{"cited_paper":"/paper/2504.19413","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:0e637475a924480324fec85e0d4eb1cfa9ee5acff9740d28051d0bf88f366b2d","observation_id":"dfab8442-935c-4bd9-8da1-e5fa4547299f","resolution":{"observed_at":"2026-07-31T12:39:14.229527Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2601.02553","last_updated":"2026-01-29T15:45:24Z","snapshot_observed_at":"2026-07-06T22:40:46.465095Z","submitted_at":"2026-01-05T21:02:49Z","title":"SimpleMem: Efficient Lifelong Memory for LLM Agents","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2601.02553","snapshot_observed_at":"2026-07-31T12:39:14.236279Z","title":"Simplemem: Efficient lifelong memory for llm agents.arXiv preprint arXiv:2601.02553, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.236279Z"},"links":{"cited_paper":"/paper/2601.02553","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:2657c322cdfd716f3d209c00fcc74d7541d4904f98f393c664a4caa42babfd0a","observation_id":"7b944d89-6043-4542-b402-5fdc09134c74","resolution":{"observed_at":"2026-07-31T12:39:14.236279Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2601.03192","last_updated":"2026-02-12T05:43:57Z","snapshot_observed_at":"2026-08-02T21:01:18.325722Z","submitted_at":"2026-01-06T17:14:50Z","title":"MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2601.03192","snapshot_observed_at":"2026-07-31T12:39:14.252240Z","title":"Memrl: Self-evolving agents via runtime reinforcement learning on episodic memory.arXiv preprint arXiv:2601.03192, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.252240Z"},"links":{"cited_paper":"/paper/2601.03192","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:0c7757c35799ab3ddc53bd7cfda344bf6b159431edd36952dcbb78c1775eeeb2","observation_id":"8c080738-af7f-4424-92e0-58e8ece96331","resolution":{"observed_at":"2026-07-31T12:39:14.252240Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2510.16079","last_updated":"2026-05-16T02:57:47Z","snapshot_observed_at":"2026-08-05T00:54:36.690304Z","submitted_at":"2025-10-17T12:03:16Z","title":"EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecycle","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2510.16079","snapshot_observed_at":"2026-07-31T12:39:14.261536Z","title":"Evolver: Self-evolving llm agents through an experience-driven lifecycle","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.261536Z"},"links":{"cited_paper":"/paper/2510.16079","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:0a0f4b18dab6ee24c9545e093dc30d5d3667b4f4e83eb92e58f45b993d9945e7","observation_id":"3457d988-770e-41bd-a42a-9751a0b777fb","resolution":{"observed_at":"2026-07-31T12:39:14.261536Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2010.03768","last_updated":"2021-03-14T22:44:38Z","snapshot_observed_at":"2026-07-06T10:02:33.297722Z","submitted_at":"2020-10-08T05:13:36Z","title":"ALFWorld: Aligning Text and Embodied Environments for Interactive Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2010.03768","snapshot_observed_at":"2026-07-31T12:39:14.271247Z","title":"Alfworld: Aligning text and embodied environments for interactive learning.arXiv preprint arXiv:2010.03768, 2020","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.271247Z"},"links":{"cited_paper":"/paper/2010.03768","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:1fb2f288c99ed0691863de2b5a650b0a196bd3feafe2854d82c5ce68820bb487","observation_id":"c26450b0-f544-4bb8-bd64-5a36da53d535","resolution":{"observed_at":"2026-07-31T12:39:14.271247Z","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":"2026-07-31T12:39:14.280917Z","title":"Webshop: Towards scalable real-world web interaction with grounded language agents","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.280917Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:cd94741bcbca3fb27a680cdc1e743635b3fedfdbb15efcbb4dfa73af1e0bcb4c","observation_id":"eb984070-f2e2-42f1-b82c-c1820b3d1a50","resolution":{"observed_at":"2026-07-31T12:39:14.280917Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.21276","last_updated":"2024-10-25T17:43:01Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-10-25T17:43:01Z","title":"GPT-4o System Card","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.21276","snapshot_observed_at":"2026-07-31T12:39:14.293899Z","title":"Gpt-4o system card.arXiv preprint arXiv:2410.21276, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.293899Z"},"links":{"cited_paper":"/paper/2410.21276","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:8f71f4cf8d4e85b13f386e760c175a55a9ba131732f5f547fe14943e7804ebc1","observation_id":"60f0886f-ce53-4528-915f-197ad3fd3282","resolution":{"observed_at":"2026-07-31T12:39:14.293899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2507.06261","last_updated":"2025-12-19T14:25:46Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2025-07-07T17:36:04Z","title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2507.06261","snapshot_observed_at":"2026-07-31T12:39:14.305492Z","title":"Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities.arXiv preprint arXiv:2507.06261, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.305492Z"},"links":{"cited_paper":"/paper/2507.06261","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:49c571949b23dc7c3ce3f90f340063664972700654fadbd7d397df50e88218b1","observation_id":"0357afa7-ce32-430f-96be-504a8e7378ec","resolution":{"observed_at":"2026-07-31T12:39:14.305492Z","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":"2026-07-06T02:11:23.670680Z","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":"2026-07-31T12:39:14.315284Z","title":"Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.315284Z"},"links":{"cited_paper":"/paper/2501.12948","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:c4890e737bfd53e8405cf0412897590f9949271de5703aa47092e6a65a9c5714","observation_id":"520365b6-03f7-4e5c-a886-4f1ad8df1b3a","resolution":{"observed_at":"2026-07-31T12:39:14.315284Z","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":"2026-07-31T12:39:14.321863Z","title":"Qwen2.5 technical report, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.321863Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:57b75d72628f4be56a82c4e2fef217ff96ebbf1b64637cea055abb2ffaa03768","observation_id":"be77f111-5239-42f9-a5c6-4784a98824d5","resolution":{"observed_at":"2026-07-31T12:39:14.321863Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.03216","last_updated":"2025-12-12T11:26:32Z","snapshot_observed_at":"2026-07-06T17:25:35.493362Z","submitted_at":"2024-02-05T17:26:49Z","title":"M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.03216","snapshot_observed_at":"2026-07-31T12:39:14.332232Z","title":"Bge m3-embedding: Multi- lingual, multi-functionality, multi-granularity text embeddings through self-knowledge distillation.arXiv preprint arXiv:2402.03216, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.332232Z"},"links":{"cited_paper":"/paper/2402.03216","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:7fb4f75f810fc95e8fccbf92015e2ec2dd7656e4d9e25aa31e5d323cf6aa06d6","observation_id":"0b948cb7-716b-45da-8ea1-e7e4f66beb57","resolution":{"observed_at":"2026-07-31T12:39:14.332232Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2409.19256","last_updated":"2024-10-02T04:01:47Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-09-28T06:20:03Z","title":"HybridFlow: A Flexible and Efficient RLHF Framework","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.19256","snapshot_observed_at":"2026-07-31T12:39:14.339493Z","title":"Hybridflow: A flexible and efficient rlhf framework.arXiv preprint arXiv: 2409.19256, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.339493Z"},"links":{"cited_paper":"/paper/2409.19256","citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:61dc07592c7a3227b20a1b43328925cdbd9bafe854269fd389df7740faa1129f","observation_id":"e9cdb622-dc33-4f2b-8cbd-fb3eb5d1a53a","resolution":{"observed_at":"2026-07-31T12:39:14.339493Z","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":"2026-07-31T12:39:14.348545Z","title":"Gon- zalez, Hao Zhang, and Ion Stoica","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.348545Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:9a89eb817a5a341883a89ccb588f94f21f01b758eb6b5fee0eb9f6880a9bc731","observation_id":"21d78a7c-e535-4953-bf95-1d6bd4d5a722","resolution":{"observed_at":"2026-07-31T12:39:14.348545Z","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":"2026-07-31T12:39:14.360834Z","title":"Milvus: A purpose-built vector data management system","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.360834Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:730dcbd2673554bcda79005c8b72888131208bf65c2f94ba1b8c6ae1da8549f1","observation_id":"a14c70bc-7d73-47bf-9fcf-596093e2b815","resolution":{"observed_at":"2026-07-31T12:39:14.360834Z","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":"2026-07-31T12:39:14.368486Z","title":"Buy Now” button. •Original State (s +):... “Price: $19.99","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-07-31T12:39:14.368486Z"},"links":{"citing_paper":"/paper/2607.28272"},"observation_digest":"sha256:0451c29a4c7142d9956c316bf338dc525512a894aff2a9116250a97d35b2d7b7","observation_id":"38923b67-8a01-45e2-ae3b-2167657286a6","resolution":{"observed_at":"2026-07-31T12:39:14.368486Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.28272","last_updated":"2026-07-30T14:25:49Z","latest_version":1,"primary_category":"cs.AI","snapshot_observed_at":"2026-08-03T00:08:22.361131Z","submitted_at":"2026-07-30T14:25:49Z","title":"MemHarness: Memory Is Reconstructed, Not Replayed"},"reference_resolution":{"displayed":38,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":38,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":38},"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":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"thesis":"As of 8 August 2026, this Paper Citation Record lists 38 of 38 outbound references and 1 inbound Pith citation observation for arXiv:2607.28272."}