{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XGEYBB5IEAFSV5X76B6S6VQLZZ","short_pith_number":"pith:XGEYBB5I","schema_version":"1.0","canonical_sha256":"b9898087a8200b2af6fff07d2f560bce65fe79a2082a2aa7557e0601f6b62b9f","source":{"kind":"arxiv","id":"2306.07929","version":2},"attestation_state":"computed","paper":{"title":"Large Language Models Are Semi-Parametric Reinforcement Learning Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Danyang Zhang, Hongshen Xu, Kai Yu, Lu Chen, Situo Zhang, Zihan Zhao","submitted_at":"2023-06-09T08:08:18Z","abstract_excerpt":"Inspired by the insights in cognitive science with respect to human memory and reasoning mechanism, a novel evolvable LLM-based (Large Language Model) agent framework is proposed as REMEMBERER. By equipping the LLM with a long-term experience memory, REMEMBERER is capable of exploiting the experiences from the past episodes even for different task goals, which excels an LLM-based agent with fixed exemplars or equipped with a transient working memory. We further introduce Reinforcement Learning with Experience Memory (RLEM) to update the memory. Thus, the whole system can learn from the experie"},"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":"2306.07929","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-09T08:08:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"137024064bd0682326b7aa157785ada3c3de7f3a1fbd711f8586914f3acf0453","abstract_canon_sha256":"c3e2cf631c4a86bf072b14490ee085a8e187c252a33a33653acb2f2e6c1a3e1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:30.103285Z","signature_b64":"FzDSWW17NnDukfFZbj1qFur5aLXAi2/kqRRUr0IxeecxbaxwMfco9E9l5BCq0QRjxRPNmK3NXe7b+4NL/CUPCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9898087a8200b2af6fff07d2f560bce65fe79a2082a2aa7557e0601f6b62b9f","last_reissued_at":"2026-07-05T07:06:30.102819Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:30.102819Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models Are Semi-Parametric Reinforcement Learning Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Danyang Zhang, Hongshen Xu, Kai Yu, Lu Chen, Situo Zhang, Zihan Zhao","submitted_at":"2023-06-09T08:08:18Z","abstract_excerpt":"Inspired by the insights in cognitive science with respect to human memory and reasoning mechanism, a novel evolvable LLM-based (Large Language Model) agent framework is proposed as REMEMBERER. By equipping the LLM with a long-term experience memory, REMEMBERER is capable of exploiting the experiences from the past episodes even for different task goals, which excels an LLM-based agent with fixed exemplars or equipped with a transient working memory. We further introduce Reinforcement Learning with Experience Memory (RLEM) to update the memory. Thus, the whole system can learn from the experie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07929","kind":"arxiv","version":2},"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/2306.07929/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":"2306.07929","created_at":"2026-07-05T07:06:30.102877+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07929v2","created_at":"2026-07-05T07:06:30.102877+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07929","created_at":"2026-07-05T07:06:30.102877+00:00"},{"alias_kind":"pith_short_12","alias_value":"XGEYBB5IEAFS","created_at":"2026-07-05T07:06:30.102877+00:00"},{"alias_kind":"pith_short_16","alias_value":"XGEYBB5IEAFSV5X7","created_at":"2026-07-05T07:06:30.102877+00:00"},{"alias_kind":"pith_short_8","alias_value":"XGEYBB5I","created_at":"2026-07-05T07:06:30.102877+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2402.02716","citing_title":"Understanding the planning of LLM agents: A survey","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17979","citing_title":"Architecture Matters More Than Scale: A Comparative Study of Retrieval and Memory Augmentation for Financial QA Under SME Compute Constraints","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17979","citing_title":"Architecture Matters More Than Scale: A Comparative Study of Retrieval and Memory Augmentation for Financial QA Under SME Compute Constraints","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ","json":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ.json","graph_json":"https://pith.science/api/pith-number/XGEYBB5IEAFSV5X76B6S6VQLZZ/graph.json","events_json":"https://pith.science/api/pith-number/XGEYBB5IEAFSV5X76B6S6VQLZZ/events.json","paper":"https://pith.science/paper/XGEYBB5I"},"agent_actions":{"view_html":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ","download_json":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ.json","view_paper":"https://pith.science/paper/XGEYBB5I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07929&json=true","fetch_graph":"https://pith.science/api/pith-number/XGEYBB5IEAFSV5X76B6S6VQLZZ/graph.json","fetch_events":"https://pith.science/api/pith-number/XGEYBB5IEAFSV5X76B6S6VQLZZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ/action/storage_attestation","attest_author":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ/action/author_attestation","sign_citation":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ/action/citation_signature","submit_replication":"https://pith.science/pith/XGEYBB5IEAFSV5X76B6S6VQLZZ/action/replication_record"}},"created_at":"2026-07-05T07:06:30.102877+00:00","updated_at":"2026-07-05T07:06:30.102877+00:00"}