{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OFHYFHOPBTZKZ6OCKLFTUBGKSL","short_pith_number":"pith:OFHYFHOP","schema_version":"1.0","canonical_sha256":"714f829dcf0cf2acf9c252cb3a04ca92e85bd1d6ced8bc641c0cc429ef6ef9e0","source":{"kind":"arxiv","id":"2606.00251","version":1},"attestation_state":"computed","paper":{"title":"Capability Self-Assessment: Teaching LLMs to Know Their Limits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haoyan Yang, Heng Huang, Jiawei Zhou, Reza Shirkavand, Shangqian Gao, Yukai Jin","submitted_at":"2026-05-29T18:32:14Z","abstract_excerpt":"The ability to recognize one's own limitations and decide whether to solve a problem or delegate is fundamental for reliable intelligent systems. Yet we show that modern large language models systematically lack this ability: across diverse model families and scales, they overestimate their competence and attempt queries they cannot solve. We refer to this ability as Capability Self-Assessment (CSA) and formulate it as a policy-learning problem, aiming to improve self-assessment while preserving the model's original capabilities. Our results show that reinforcement learning teaches CSA effecti"},"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":"2606.00251","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-29T18:32:14Z","cross_cats_sorted":[],"title_canon_sha256":"4cb4b821ec8f06e080c920e15874735ff5549d8a61a599a8ca75afce2c452644","abstract_canon_sha256":"e613205eb7fc444207956f2d552d34a35ef06b68bbfac03f9a4edbceed5fe6de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:22.788431Z","signature_b64":"6OP9QJ4hUHRY3JMrjtJROO7IqdjokRZyBHZeBUXP4iTJY8AhlPP9RsIJ2y/QFjvjUOR5ccRuOzrEnNB/kR0fBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"714f829dcf0cf2acf9c252cb3a04ca92e85bd1d6ced8bc641c0cc429ef6ef9e0","last_reissued_at":"2026-06-02T01:03:22.788067Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:22.788067Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Capability Self-Assessment: Teaching LLMs to Know Their Limits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haoyan Yang, Heng Huang, Jiawei Zhou, Reza Shirkavand, Shangqian Gao, Yukai Jin","submitted_at":"2026-05-29T18:32:14Z","abstract_excerpt":"The ability to recognize one's own limitations and decide whether to solve a problem or delegate is fundamental for reliable intelligent systems. Yet we show that modern large language models systematically lack this ability: across diverse model families and scales, they overestimate their competence and attempt queries they cannot solve. We refer to this ability as Capability Self-Assessment (CSA) and formulate it as a policy-learning problem, aiming to improve self-assessment while preserving the model's original capabilities. Our results show that reinforcement learning teaches CSA effecti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00251","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/2606.00251/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":"2606.00251","created_at":"2026-06-02T01:03:22.788126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00251v1","created_at":"2026-06-02T01:03:22.788126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00251","created_at":"2026-06-02T01:03:22.788126+00:00"},{"alias_kind":"pith_short_12","alias_value":"OFHYFHOPBTZK","created_at":"2026-06-02T01:03:22.788126+00:00"},{"alias_kind":"pith_short_16","alias_value":"OFHYFHOPBTZKZ6OC","created_at":"2026-06-02T01:03:22.788126+00:00"},{"alias_kind":"pith_short_8","alias_value":"OFHYFHOP","created_at":"2026-06-02T01:03:22.788126+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/OFHYFHOPBTZKZ6OCKLFTUBGKSL","json":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL.json","graph_json":"https://pith.science/api/pith-number/OFHYFHOPBTZKZ6OCKLFTUBGKSL/graph.json","events_json":"https://pith.science/api/pith-number/OFHYFHOPBTZKZ6OCKLFTUBGKSL/events.json","paper":"https://pith.science/paper/OFHYFHOP"},"agent_actions":{"view_html":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL","download_json":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL.json","view_paper":"https://pith.science/paper/OFHYFHOP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00251&json=true","fetch_graph":"https://pith.science/api/pith-number/OFHYFHOPBTZKZ6OCKLFTUBGKSL/graph.json","fetch_events":"https://pith.science/api/pith-number/OFHYFHOPBTZKZ6OCKLFTUBGKSL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/action/storage_attestation","attest_author":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/action/author_attestation","sign_citation":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/action/citation_signature","submit_replication":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/action/replication_record"}},"created_at":"2026-06-02T01:03:22.788126+00:00","updated_at":"2026-06-02T01:03:22.788126+00:00"}