{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PTNQXGGL43Y6PFPS3UZ3LO6HT7","short_pith_number":"pith:PTNQXGGL","schema_version":"1.0","canonical_sha256":"7cdb0b98cbe6f1e795f2dd33b5bbc79ff67ce63ed8b4f9972b341aee061acaf5","source":{"kind":"arxiv","id":"2402.17574","version":3},"attestation_state":"computed","paper":{"title":"Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Guiyang Hou, Hai Wu, Ke Tang, Mengna Wang, Peng Li, Weiming Lu, Wenqi Zhang, Yongliang Shen, Yueting Zhuang, Zeqi Tan","submitted_at":"2024-02-27T15:09:20Z","abstract_excerpt":"Large Language Models (LLMs) exhibit robust problem-solving capabilities for diverse tasks. However, most LLM-based agents are designed as specific task solvers with sophisticated prompt engineering, rather than agents capable of learning and evolving through interactions. These task solvers necessitate manually crafted prompts to inform task rules and regulate LLM behaviors, inherently incapacitating to address complex dynamic scenarios e.g., large interactive games. In light of this, we propose Agent-Pro: an LLM-based Agent with Policy-level Reflection and Optimization that can learn a wealt"},"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":"2402.17574","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-27T15:09:20Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ce7c26f137d63a08af78b5df62fbc7bbeda5b657c3ebcf3b12f60075f4212b32","abstract_canon_sha256":"4e33f656603aad176cdfb5d26506b101226b9c2a984a79ac945f94a94c63b2d2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:39.053374Z","signature_b64":"mFRSAfhjZdggUrkV+PqvkxpN83fZdgb6cKu9avwBGTThpD3vJkHAx7xo2YqEZNiRnzjL52yfY9jjS3Q2JAbdAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7cdb0b98cbe6f1e795f2dd33b5bbc79ff67ce63ed8b4f9972b341aee061acaf5","last_reissued_at":"2026-07-05T08:28:39.052915Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:39.052915Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Guiyang Hou, Hai Wu, Ke Tang, Mengna Wang, Peng Li, Weiming Lu, Wenqi Zhang, Yongliang Shen, Yueting Zhuang, Zeqi Tan","submitted_at":"2024-02-27T15:09:20Z","abstract_excerpt":"Large Language Models (LLMs) exhibit robust problem-solving capabilities for diverse tasks. However, most LLM-based agents are designed as specific task solvers with sophisticated prompt engineering, rather than agents capable of learning and evolving through interactions. These task solvers necessitate manually crafted prompts to inform task rules and regulate LLM behaviors, inherently incapacitating to address complex dynamic scenarios e.g., large interactive games. In light of this, we propose Agent-Pro: an LLM-based Agent with Policy-level Reflection and Optimization that can learn a wealt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.17574","kind":"arxiv","version":3},"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/2402.17574/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":"2402.17574","created_at":"2026-07-05T08:28:39.052975+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.17574v3","created_at":"2026-07-05T08:28:39.052975+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.17574","created_at":"2026-07-05T08:28:39.052975+00:00"},{"alias_kind":"pith_short_12","alias_value":"PTNQXGGL43Y6","created_at":"2026-07-05T08:28:39.052975+00:00"},{"alias_kind":"pith_short_16","alias_value":"PTNQXGGL43Y6PFPS","created_at":"2026-07-05T08:28:39.052975+00:00"},{"alias_kind":"pith_short_8","alias_value":"PTNQXGGL","created_at":"2026-07-05T08:28:39.052975+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24428","citing_title":"Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01581","citing_title":"Agent System Operations: Categorization, Challenges, and Future Directions","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13388","citing_title":"Reflection of Episodes: Learning to Play Game from Expert and Self Experiences","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05974","citing_title":"PragLocker: Protecting Agent Intellectual Property in Untrusted Deployments via Non-Portable Prompts","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09686","citing_title":"Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models","ref_index":190,"is_internal_anchor":false},{"citing_arxiv_id":"2603.29908","citing_title":"C-TRAIL: A Commonsense World Framework for Trajectory Planning in Autonomous Driving","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05974","citing_title":"PragLocker: Protecting Agent Intellectual Property in Untrusted Deployments via Non-Portable Prompts","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09285","citing_title":"SAGE: A Service Agent Graph-guided Evaluation Benchmark","ref_index":66,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7","json":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7.json","graph_json":"https://pith.science/api/pith-number/PTNQXGGL43Y6PFPS3UZ3LO6HT7/graph.json","events_json":"https://pith.science/api/pith-number/PTNQXGGL43Y6PFPS3UZ3LO6HT7/events.json","paper":"https://pith.science/paper/PTNQXGGL"},"agent_actions":{"view_html":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7","download_json":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7.json","view_paper":"https://pith.science/paper/PTNQXGGL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.17574&json=true","fetch_graph":"https://pith.science/api/pith-number/PTNQXGGL43Y6PFPS3UZ3LO6HT7/graph.json","fetch_events":"https://pith.science/api/pith-number/PTNQXGGL43Y6PFPS3UZ3LO6HT7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7/action/storage_attestation","attest_author":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7/action/author_attestation","sign_citation":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7/action/citation_signature","submit_replication":"https://pith.science/pith/PTNQXGGL43Y6PFPS3UZ3LO6HT7/action/replication_record"}},"created_at":"2026-07-05T08:28:39.052975+00:00","updated_at":"2026-07-05T08:28:39.052975+00:00"}