{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QY3GIVS3C26GE2OYMUB3LDHCOQ","short_pith_number":"pith:QY3GIVS3","schema_version":"1.0","canonical_sha256":"863664565b16bc6269d86503b58ce27407cdc0a0b749dbfb13d157565c7d6630","source":{"kind":"arxiv","id":"2410.15665","version":4},"attestation_state":"computed","paper":{"title":"Long Term Memory: The Foundation of AI Self-Evolution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Feng Li, Han Zhao, Jiahao Qiu, Jiaying Wang, Jun Shao, Mengdi Wang, Mengyue Wu, Shihao Xu, Shu Zhang, Tianqiao Chen, Weiling Chen, Weizhi Ma, Xavier Tang, Xun Jiang, Yize Chen","submitted_at":"2024-10-21T06:09:30Z","abstract_excerpt":"Large language models (LLMs) like GPTs, trained on vast datasets, have demonstrated impressive capabilities in language understanding, reasoning, and planning, achieving human-level performance in various tasks. Most studies focus on enhancing these models by training on ever-larger datasets to build more powerful foundation models. While training stronger models is important, enabling models to evolve during inference is equally crucial, a process we refer to as AI self-evolution. Unlike large-scale training, self-evolution may rely on limited data or interactions. Inspired by the columnar or"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.15665","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-10-21T06:09:30Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"683e91a3ad90ebfcb5b3a2867038ed8deca68b1e986a9722196dcb05779ca5ca","abstract_canon_sha256":"84d764a7ca2f02ee1426e57c8cd6d54ac90765241df0712dc9dd4ce26b599205"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:01:27.778457Z","signature_b64":"xJaVnmsXWmJYPg8r3pYNjveKsal4EOF6wWw3fuTvUchu75pbDHbxndDdqyakZQxgd+OG04+oBj2o1Mi2RlKICA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"863664565b16bc6269d86503b58ce27407cdc0a0b749dbfb13d157565c7d6630","last_reissued_at":"2026-07-05T11:01:27.777754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:01:27.777754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Long Term Memory: The Foundation of AI Self-Evolution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Feng Li, Han Zhao, Jiahao Qiu, Jiaying Wang, Jun Shao, Mengdi Wang, Mengyue Wu, Shihao Xu, Shu Zhang, Tianqiao Chen, Weiling Chen, Weizhi Ma, Xavier Tang, Xun Jiang, Yize Chen","submitted_at":"2024-10-21T06:09:30Z","abstract_excerpt":"Large language models (LLMs) like GPTs, trained on vast datasets, have demonstrated impressive capabilities in language understanding, reasoning, and planning, achieving human-level performance in various tasks. Most studies focus on enhancing these models by training on ever-larger datasets to build more powerful foundation models. While training stronger models is important, enabling models to evolve during inference is equally crucial, a process we refer to as AI self-evolution. Unlike large-scale training, self-evolution may rely on limited data or interactions. Inspired by the columnar or"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.15665","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.15665/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.15665","created_at":"2026-07-05T11:01:27.777859+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.15665v4","created_at":"2026-07-05T11:01:27.777859+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.15665","created_at":"2026-07-05T11:01:27.777859+00:00"},{"alias_kind":"pith_short_12","alias_value":"QY3GIVS3C26G","created_at":"2026-07-05T11:01:27.777859+00:00"},{"alias_kind":"pith_short_16","alias_value":"QY3GIVS3C26GE2OY","created_at":"2026-07-05T11:01:27.777859+00:00"},{"alias_kind":"pith_short_8","alias_value":"QY3GIVS3","created_at":"2026-07-05T11:01:27.777859+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03014","citing_title":"MOSAIC: Efficient Mixture-of-Agent Scheduling via Adaptive Aggregation and Inference Concurrency","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2504.15965","citing_title":"From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21046","citing_title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ","json":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ.json","graph_json":"https://pith.science/api/pith-number/QY3GIVS3C26GE2OYMUB3LDHCOQ/graph.json","events_json":"https://pith.science/api/pith-number/QY3GIVS3C26GE2OYMUB3LDHCOQ/events.json","paper":"https://pith.science/paper/QY3GIVS3"},"agent_actions":{"view_html":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ","download_json":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ.json","view_paper":"https://pith.science/paper/QY3GIVS3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.15665&json=true","fetch_graph":"https://pith.science/api/pith-number/QY3GIVS3C26GE2OYMUB3LDHCOQ/graph.json","fetch_events":"https://pith.science/api/pith-number/QY3GIVS3C26GE2OYMUB3LDHCOQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ/action/storage_attestation","attest_author":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ/action/author_attestation","sign_citation":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ/action/citation_signature","submit_replication":"https://pith.science/pith/QY3GIVS3C26GE2OYMUB3LDHCOQ/action/replication_record"}},"created_at":"2026-07-05T11:01:27.777859+00:00","updated_at":"2026-07-05T11:01:27.777859+00:00"}