{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JN25CBIEBKAQ2CBTIMCBCOHZ2B","short_pith_number":"pith:JN25CBIE","schema_version":"1.0","canonical_sha256":"4b75d105040a810d083343041138f9d062d6b34e586a9b99545c6b54d6a127fb","source":{"kind":"arxiv","id":"2506.05109","version":1},"attestation_state":"computed","paper":{"title":"Truly Self-Improving Agents Require Intrinsic Metacognitive Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Mihaela van der Schaar, Tennison Liu","submitted_at":"2025-06-05T14:53:35Z","abstract_excerpt":"Self-improving agents aim to continuously acquire new capabilities with minimal supervision. However, current approaches face two key limitations: their self-improvement processes are often rigid, fail to generalize across tasks domains, and struggle to scale with increasing agent capabilities. We argue that effective self-improvement requires intrinsic metacognitive learning, defined as an agent's intrinsic ability to actively evaluate, reflect on, and adapt its own learning processes. Drawing inspiration from human metacognition, we introduce a formal framework comprising three components: m"},"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":"2506.05109","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-05T14:53:35Z","cross_cats_sorted":[],"title_canon_sha256":"aafb800d6cd1b984ad2c1e0662f162c41dd9329348432773cdd8ec09a2ca7c98","abstract_canon_sha256":"be8345f401a2fd9105a5528b7a8fde336fdaa8b0eb330aedb2545b41da691438"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:42.488726Z","signature_b64":"5/24K2LFZr+1GaGppfEL66pBizcx+ju5mOc6q1KJCtu0lZoq/yn7Dkw+lYL+i7BCHbRfRcWq4qmOMZgofwY2Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b75d105040a810d083343041138f9d062d6b34e586a9b99545c6b54d6a127fb","last_reissued_at":"2026-07-05T11:16:42.488193Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:42.488193Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Truly Self-Improving Agents Require Intrinsic Metacognitive Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Mihaela van der Schaar, Tennison Liu","submitted_at":"2025-06-05T14:53:35Z","abstract_excerpt":"Self-improving agents aim to continuously acquire new capabilities with minimal supervision. However, current approaches face two key limitations: their self-improvement processes are often rigid, fail to generalize across tasks domains, and struggle to scale with increasing agent capabilities. We argue that effective self-improvement requires intrinsic metacognitive learning, defined as an agent's intrinsic ability to actively evaluate, reflect on, and adapt its own learning processes. Drawing inspiration from human metacognition, we introduce a formal framework comprising three components: m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05109","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/2506.05109/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":"2506.05109","created_at":"2026-07-05T11:16:42.488262+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.05109v1","created_at":"2026-07-05T11:16:42.488262+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05109","created_at":"2026-07-05T11:16:42.488262+00:00"},{"alias_kind":"pith_short_12","alias_value":"JN25CBIEBKAQ","created_at":"2026-07-05T11:16:42.488262+00:00"},{"alias_kind":"pith_short_16","alias_value":"JN25CBIEBKAQ2CBT","created_at":"2026-07-05T11:16:42.488262+00:00"},{"alias_kind":"pith_short_8","alias_value":"JN25CBIE","created_at":"2026-07-05T11:16:42.488262+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07663","citing_title":"Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops","ref_index":61,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23127","citing_title":"Managing Procedural Memory in LLM Agents: Control, Adaptation, and Evaluation","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B","json":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B.json","graph_json":"https://pith.science/api/pith-number/JN25CBIEBKAQ2CBTIMCBCOHZ2B/graph.json","events_json":"https://pith.science/api/pith-number/JN25CBIEBKAQ2CBTIMCBCOHZ2B/events.json","paper":"https://pith.science/paper/JN25CBIE"},"agent_actions":{"view_html":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B","download_json":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B.json","view_paper":"https://pith.science/paper/JN25CBIE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.05109&json=true","fetch_graph":"https://pith.science/api/pith-number/JN25CBIEBKAQ2CBTIMCBCOHZ2B/graph.json","fetch_events":"https://pith.science/api/pith-number/JN25CBIEBKAQ2CBTIMCBCOHZ2B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B/action/storage_attestation","attest_author":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B/action/author_attestation","sign_citation":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B/action/citation_signature","submit_replication":"https://pith.science/pith/JN25CBIEBKAQ2CBTIMCBCOHZ2B/action/replication_record"}},"created_at":"2026-07-05T11:16:42.488262+00:00","updated_at":"2026-07-05T11:16:42.488262+00:00"}