{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DKVDHXPMNPQINWXRN4F3AVS3JP","short_pith_number":"pith:DKVDHXPM","schema_version":"1.0","canonical_sha256":"1aaa33ddec6be086daf16f0bb0565b4be9c8422a0404e61a6f6aa58af364880d","source":{"kind":"arxiv","id":"2605.30159","version":1},"attestation_state":"computed","paper":{"title":"Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Feng Liu, Hong Wang, Jingren Hou, Ruiyi Ding, Wei Xia, Wence Ji, Yeqiu Chen, Yongkang Yang, Zhezheng Hao, Ziyan Liu","submitted_at":"2026-05-28T16:17:19Z","abstract_excerpt":"Memory-augmented LLM agents tackle complex long-horizon tasks by recursively summarizing interaction trajectories into compact memory. However, existing approaches typically train these memory policies using outcome-based reinforcement learning, failing to localize where intermediate memory quality degrades. As interactions unfold, ambiguous recursive summaries progressively discard task-relevant information and introduce semantic noise. This exacerbates belief deviation, obscuring the agent's estimate of the latent task state and ultimately derailing long-horizon reasoning. We therefore argue"},"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":"2605.30159","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T16:17:19Z","cross_cats_sorted":[],"title_canon_sha256":"fdf579b53aa9776735b12ea1ec52670e7c0fac960e08feae7ab00d965c2e660e","abstract_canon_sha256":"1a7882401d9b732c7e2b3e95c1f68d420fa2649be3ab7d7cbabb8eeda56094c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:06:11.370075Z","signature_b64":"jX+rQiY9Lv0TjLFEsTyNwbfOOT7c33Hnvr+jNCWPxcdZVaOy/htOx7rMh2CkDto3XWMJiAfQj1bZ8bWKbpLrCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1aaa33ddec6be086daf16f0bb0565b4be9c8422a0404e61a6f6aa58af364880d","last_reissued_at":"2026-05-29T02:06:11.369605Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:06:11.369605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Feng Liu, Hong Wang, Jingren Hou, Ruiyi Ding, Wei Xia, Wence Ji, Yeqiu Chen, Yongkang Yang, Zhezheng Hao, Ziyan Liu","submitted_at":"2026-05-28T16:17:19Z","abstract_excerpt":"Memory-augmented LLM agents tackle complex long-horizon tasks by recursively summarizing interaction trajectories into compact memory. However, existing approaches typically train these memory policies using outcome-based reinforcement learning, failing to localize where intermediate memory quality degrades. As interactions unfold, ambiguous recursive summaries progressively discard task-relevant information and introduce semantic noise. This exacerbates belief deviation, obscuring the agent's estimate of the latent task state and ultimately derailing long-horizon reasoning. We therefore argue"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30159","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/2605.30159/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":"2605.30159","created_at":"2026-05-29T02:06:11.369685+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30159v1","created_at":"2026-05-29T02:06:11.369685+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30159","created_at":"2026-05-29T02:06:11.369685+00:00"},{"alias_kind":"pith_short_12","alias_value":"DKVDHXPMNPQI","created_at":"2026-05-29T02:06:11.369685+00:00"},{"alias_kind":"pith_short_16","alias_value":"DKVDHXPMNPQINWXR","created_at":"2026-05-29T02:06:11.369685+00:00"},{"alias_kind":"pith_short_8","alias_value":"DKVDHXPM","created_at":"2026-05-29T02:06:11.369685+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/DKVDHXPMNPQINWXRN4F3AVS3JP","json":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP.json","graph_json":"https://pith.science/api/pith-number/DKVDHXPMNPQINWXRN4F3AVS3JP/graph.json","events_json":"https://pith.science/api/pith-number/DKVDHXPMNPQINWXRN4F3AVS3JP/events.json","paper":"https://pith.science/paper/DKVDHXPM"},"agent_actions":{"view_html":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP","download_json":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP.json","view_paper":"https://pith.science/paper/DKVDHXPM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30159&json=true","fetch_graph":"https://pith.science/api/pith-number/DKVDHXPMNPQINWXRN4F3AVS3JP/graph.json","fetch_events":"https://pith.science/api/pith-number/DKVDHXPMNPQINWXRN4F3AVS3JP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP/action/storage_attestation","attest_author":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP/action/author_attestation","sign_citation":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP/action/citation_signature","submit_replication":"https://pith.science/pith/DKVDHXPMNPQINWXRN4F3AVS3JP/action/replication_record"}},"created_at":"2026-05-29T02:06:11.369685+00:00","updated_at":"2026-05-29T02:06:11.369685+00:00"}