{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MJNLBHKNIQZAS4LHX3DLU6OO3E","short_pith_number":"pith:MJNLBHKN","schema_version":"1.0","canonical_sha256":"625ab09d4d4432097167bec6ba79ced923466ffde4074b16a48744367545bf1b","source":{"kind":"arxiv","id":"2605.21240","version":1},"attestation_state":"computed","paper":{"title":"APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bryan Hooi, Jiashuo Yang, Shizun Wang, Yibo Li, Yuan Sui, Yufei He, Zhiyuan Hu, Zhi Zheng","submitted_at":"2026-05-20T14:29:27Z","abstract_excerpt":"LLM agents have shown strong performance across a wide range of complex tasks, including interactive environments that require long-horizon decision making. But these agents cannot learn on the fly at test time. Self-evolving agents address this by accumulating memory and reflection across episodes rather than requiring model-weight updates. However, these agents often suffer from exploration collapse: as memory grows, behavior concentrates around familiar high-reward routines, reducing the chance of discovering better alternatives. To address this problem, we propose Autonomous Policy EXplora"},"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":"2605.21240","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:29:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8c5f63ba2fec572c0cd721197dc207c85ec81fba86080e874b0144dbdaba38b5","abstract_canon_sha256":"7c4c29552d2761c7a62d3ebc90ce5509d08ca58863f58fbc0b06ad90526807e5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:45.307929Z","signature_b64":"r6DqxPaycXJLJDgs+AFD7wKwKpEh155hafz43HD6gfKOJ6qxKplzgZEmxbAs+dDDGO4ubv+16hC3/NcfSCKIDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"625ab09d4d4432097167bec6ba79ced923466ffde4074b16a48744367545bf1b","last_reissued_at":"2026-05-21T01:05:45.307350Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:45.307350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bryan Hooi, Jiashuo Yang, Shizun Wang, Yibo Li, Yuan Sui, Yufei He, Zhiyuan Hu, Zhi Zheng","submitted_at":"2026-05-20T14:29:27Z","abstract_excerpt":"LLM agents have shown strong performance across a wide range of complex tasks, including interactive environments that require long-horizon decision making. But these agents cannot learn on the fly at test time. Self-evolving agents address this by accumulating memory and reflection across episodes rather than requiring model-weight updates. However, these agents often suffer from exploration collapse: as memory grows, behavior concentrates around familiar high-reward routines, reducing the chance of discovering better alternatives. To address this problem, we propose Autonomous Policy EXplora"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21240","kind":"arxiv","version":1},"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/2605.21240/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":"2605.21240","created_at":"2026-05-21T01:05:45.307436+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21240v1","created_at":"2026-05-21T01:05:45.307436+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21240","created_at":"2026-05-21T01:05:45.307436+00:00"},{"alias_kind":"pith_short_12","alias_value":"MJNLBHKNIQZA","created_at":"2026-05-21T01:05:45.307436+00:00"},{"alias_kind":"pith_short_16","alias_value":"MJNLBHKNIQZAS4LH","created_at":"2026-05-21T01:05:45.307436+00:00"},{"alias_kind":"pith_short_8","alias_value":"MJNLBHKN","created_at":"2026-05-21T01:05:45.307436+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E","json":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E.json","graph_json":"https://pith.science/api/pith-number/MJNLBHKNIQZAS4LHX3DLU6OO3E/graph.json","events_json":"https://pith.science/api/pith-number/MJNLBHKNIQZAS4LHX3DLU6OO3E/events.json","paper":"https://pith.science/paper/MJNLBHKN"},"agent_actions":{"view_html":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E","download_json":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E.json","view_paper":"https://pith.science/paper/MJNLBHKN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21240&json=true","fetch_graph":"https://pith.science/api/pith-number/MJNLBHKNIQZAS4LHX3DLU6OO3E/graph.json","fetch_events":"https://pith.science/api/pith-number/MJNLBHKNIQZAS4LHX3DLU6OO3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E/action/storage_attestation","attest_author":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E/action/author_attestation","sign_citation":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E/action/citation_signature","submit_replication":"https://pith.science/pith/MJNLBHKNIQZAS4LHX3DLU6OO3E/action/replication_record"}},"created_at":"2026-05-21T01:05:45.307436+00:00","updated_at":"2026-05-21T01:05:45.307436+00:00"}