{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:EFGOQGWQBVIWVZWNPFASUUA6RN","short_pith_number":"pith:EFGOQGWQ","schema_version":"1.0","canonical_sha256":"214ce81ad00d516ae6cd79412a501e8b71f2b5e54e2a48340ccaaf634396e86e","source":{"kind":"arxiv","id":"2309.02427","version":3},"attestation_state":"computed","paper":{"title":"Cognitive Architectures for Language Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CoALA structures language agents with modular memory components, a structured action space, and a generalized decision-making process drawn from cognitive science.","cross_cats":["cs.CL","cs.LG","cs.SC"],"primary_cat":"cs.AI","authors_text":"Karthik Narasimhan, Shunyu Yao, Theodore R. Sumers, Thomas L. Griffiths","submitted_at":"2023-09-05T17:56:20Z","abstract_excerpt":"Recent efforts have augmented large language models (LLMs) with external resources (e.g., the Internet) or internal control flows (e.g., prompt chaining) for tasks requiring grounding or reasoning, leading to a new class of language agents. While these agents have achieved substantial empirical success, we lack a systematic framework to organize existing agents and plan future developments. In this paper, we draw on the rich history of cognitive science and symbolic artificial intelligence to propose Cognitive Architectures for Language Agents (CoALA). CoALA describes a language agent with mod"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2309.02427","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-09-05T17:56:20Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SC"],"title_canon_sha256":"8a85a0b59f7d58386b473f0e0ff4571994e49864b784d3967212a867e6a17fbd","abstract_canon_sha256":"86d0b6bd5b1490ba081f23a157aff529ac85d432f794093827d5fc65e05eba9e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:46.833421Z","signature_b64":"uqeu/QHBN6iekJgaMJw1pzApp6rZEQLy9d6mBWRpXttftUtWFvrV/Xv4bmPaVLzha799UT8m9OlBuFt3kX2fDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"214ce81ad00d516ae6cd79412a501e8b71f2b5e54e2a48340ccaaf634396e86e","last_reissued_at":"2026-05-17T23:38:46.832897Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:46.832897Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cognitive Architectures for Language Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CoALA structures language agents with modular memory components, a structured action space, and a generalized decision-making process drawn from cognitive science.","cross_cats":["cs.CL","cs.LG","cs.SC"],"primary_cat":"cs.AI","authors_text":"Karthik Narasimhan, Shunyu Yao, Theodore R. Sumers, Thomas L. Griffiths","submitted_at":"2023-09-05T17:56:20Z","abstract_excerpt":"Recent efforts have augmented large language models (LLMs) with external resources (e.g., the Internet) or internal control flows (e.g., prompt chaining) for tasks requiring grounding or reasoning, leading to a new class of language agents. While these agents have achieved substantial empirical success, we lack a systematic framework to organize existing agents and plan future developments. In this paper, we draw on the rich history of cognitive science and symbolic artificial intelligence to propose Cognitive Architectures for Language Agents (CoALA). CoALA describes a language agent with mod"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CoALA describes a language agent with modular memory components, a structured action space to interact with internal memory and external environments, and a generalized decision-making process to choose actions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That principles from cognitive science and symbolic AI can be directly transferred to LLM-based agents without major adaptation or loss of the benefits that make LLMs effective in the first place.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CoALA is a modular cognitive architecture for language agents that organizes memory components, action spaces for internal and external interaction, and a generalized decision-making loop to support more systematic development of capable agents.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CoALA structures language agents with modular memory components, a structured action space, and a generalized decision-making process drawn from cognitive science.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5c622eeb3505c4f5f52273d0b223c5b330154dcc2318290fb9a9569552573996"},"source":{"id":"2309.02427","kind":"arxiv","version":3},"verdict":{"id":"8334f6db-e9df-414f-a921-7b9030b723ad","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T19:30:13.643090Z","strongest_claim":"CoALA describes a language agent with modular memory components, a structured action space to interact with internal memory and external environments, and a generalized decision-making process to choose actions.","one_line_summary":"CoALA is a modular cognitive architecture for language agents that organizes memory components, action spaces for internal and external interaction, and a generalized decision-making loop to support more systematic development of capable agents.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That principles from cognitive science and symbolic AI can be directly transferred to LLM-based agents without major adaptation or loss of the benefits that make LLMs effective in the first place.","pith_extraction_headline":"CoALA structures language agents with modular memory components, a structured action space, and a generalized decision-making process drawn from cognitive science."},"references":{"count":96,"sample":[{"doi":"","year":null,"title":"Do As I Can, Not As I Say: Grounding Language in Robotic Affordances","work_id":"037320f1-b0a9-4cbe-a639-bfb25409ce71","ref_index":1,"cited_arxiv_id":"2204.01691","is_internal_anchor":true},{"doi":"","year":2022,"title":"J. Andreas. Language models as agent models. InFindings of the Association for Computational Linguistics: EMNLP 2022, pages 5769–5779,","work_id":"b64095f4-b7fb-4cda-983b-a49197c74937","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"19 Published in Transactions on Machine Learning Research (02/2024) A. D. Baddeley and G. Hitch. Working memory. InPsychology of Learning and Motivation, volume 8, pages 47–89. Elsevier,","work_id":"bb3dec28-069c-45e9-931a-9c942950bb50","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","ref_index":4,"cited_arxiv_id":"2212.08073","is_internal_anchor":true},{"doi":"","year":null,"title":"Z., Rae, J., Wierstra, D., and Hass- abis, D","work_id":"9aff3f85-02c1-4ef6-90e8-7d051bc14cf0","ref_index":5,"cited_arxiv_id":"1606.04460","is_internal_anchor":true}],"resolved_work":96,"snapshot_sha256":"fe4977d40fb76bf42fa878092ca6eaf5567ebc0040c3cf3a5837a61df14f39ce","internal_anchors":45},"formal_canon":{"evidence_count":3,"snapshot_sha256":"7f209224f593ae98784b205a605b89c3f75842dc643bc3f52c02be1a4462b17d"},"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":"2309.02427","created_at":"2026-05-17T23:38:46.832982+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.02427v3","created_at":"2026-05-17T23:38:46.832982+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.02427","created_at":"2026-05-17T23:38:46.832982+00:00"},{"alias_kind":"pith_short_12","alias_value":"EFGOQGWQBVIW","created_at":"2026-05-18T12:33:33.725879+00:00"},{"alias_kind":"pith_short_16","alias_value":"EFGOQGWQBVIWVZWN","created_at":"2026-05-18T12:33:33.725879+00:00"},{"alias_kind":"pith_short_8","alias_value":"EFGOQGWQ","created_at":"2026-05-18T12:33:33.725879+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":26,"internal_anchor_count":26,"sample":[{"citing_arxiv_id":"2604.02688","citing_title":"MatClaw: An Autonomous Code-First LLM Agent for End-to-End Materials Exploration","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22643","citing_title":"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22643","citing_title":"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2507.21035","citing_title":"GenoMAS: A Multi-Agent Framework for Scientific Discovery via Code-Driven Gene Expression Analysis","ref_index":107,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21463","citing_title":"Mem-$\\pi$: Adaptive Memory through Learning When and What to Generate","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20724","citing_title":"CALMem : Application-Layer Dual Memory for Conversational AI","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18854","citing_title":"Evaluating Memory Condensation Strategies for Coding Agents in Data-Driven Scientific Discovery","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17625","citing_title":"Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2401.05459","citing_title":"Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Security","ref_index":226,"is_internal_anchor":true},{"citing_arxiv_id":"2402.05070","citing_title":"A Roadmap to Pluralistic Alignment","ref_index":269,"is_internal_anchor":true},{"citing_arxiv_id":"2501.09686","citing_title":"Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models","ref_index":139,"is_internal_anchor":true},{"citing_arxiv_id":"2410.07283","citing_title":"Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems","ref_index":52,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15184","citing_title":"Is Grep All You Need? How Agent Harnesses Reshape Agentic Search","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13625","citing_title":"How to Interpret Agent Behavior","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02688","citing_title":"MatClaw: An Autonomous Code-First LLM Agent for End-to-End Materials Exploration","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2410.23218","citing_title":"OS-ATLAS: A Foundation Action Model for Generalist GUI Agents","ref_index":72,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26622","citing_title":"OCR-Memory: Optical Context Retrieval for Long-Horizon Agent Memory","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2402.01680","citing_title":"Large Language Model based Multi-Agents: A Survey of Progress and Challenges","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05716","citing_title":"More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22085","citing_title":"Memanto: Typed Semantic Memory with Information-Theoretic Retrieval for Long-Horizon Agents","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2309.07864","citing_title":"The Rise and Potential of Large Language Model Based Agents: A Survey","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10352","citing_title":"ClawVM: Harness-Managed Virtual Memory for Stateful Tool-Using LLM Agents","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2604.11364","citing_title":"The Missing Knowledge Layer in Cognitive Architectures for AI Agents","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07595","citing_title":"ROZA Graphs: Self-Improving Near-Deterministic RAG through Evidence-Centric Feedback","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2406.12045","citing_title":"$\\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":3,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN","json":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN.json","graph_json":"https://pith.science/api/pith-number/EFGOQGWQBVIWVZWNPFASUUA6RN/graph.json","events_json":"https://pith.science/api/pith-number/EFGOQGWQBVIWVZWNPFASUUA6RN/events.json","paper":"https://pith.science/paper/EFGOQGWQ"},"agent_actions":{"view_html":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN","download_json":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN.json","view_paper":"https://pith.science/paper/EFGOQGWQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.02427&json=true","fetch_graph":"https://pith.science/api/pith-number/EFGOQGWQBVIWVZWNPFASUUA6RN/graph.json","fetch_events":"https://pith.science/api/pith-number/EFGOQGWQBVIWVZWNPFASUUA6RN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN/action/storage_attestation","attest_author":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN/action/author_attestation","sign_citation":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN/action/citation_signature","submit_replication":"https://pith.science/pith/EFGOQGWQBVIWVZWNPFASUUA6RN/action/replication_record"}},"created_at":"2026-05-17T23:38:46.832982+00:00","updated_at":"2026-05-17T23:38:46.832982+00:00"}