{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OFHYFHOPBTZKZ6OCKLFTUBGKSL","short_pith_number":"pith:OFHYFHOP","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"},"canonical_sha256":"714f829dcf0cf2acf9c252cb3a04ca92e85bd1d6ced8bc641c0cc429ef6ef9e0","source":{"kind":"arxiv","id":"2606.00251","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00251","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00251v1","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00251","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"pith_short_12","alias_value":"OFHYFHOPBTZK","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"pith_short_16","alias_value":"OFHYFHOPBTZKZ6OC","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"pith_short_8","alias_value":"OFHYFHOP","created_at":"2026-06-02T01:03:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OFHYFHOPBTZKZ6OCKLFTUBGKSL","target":"record","payload":{"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"},"canonical_sha256":"714f829dcf0cf2acf9c252cb3a04ca92e85bd1d6ced8bc641c0cc429ef6ef9e0","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"},"source_kind":"arxiv","source_id":"2606.00251","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2YC353lu3C1VEDLZRLRe3+fjkon6hh22uGJgp9SgORKVaLsu9AiigxVa2vxPDpTWcUAlsKr0LppKTjAbYvNdDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T11:58:36.024382Z"},"content_sha256":"4fe6adcaeedf178c525e0f499215bb8518fbbb5a827b93392e53b78b8db5e9ca","schema_version":"1.0","event_id":"sha256:4fe6adcaeedf178c525e0f499215bb8518fbbb5a827b93392e53b78b8db5e9ca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OFHYFHOPBTZKZ6OCKLFTUBGKSL","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3BnUDdDSZhT0WjpykmRjGz6AjUYPXxsQHBxaJd498Wo4lZ1rfHdH68zRTBCzWsHOJAXMr+fxKWwgDMSnkDv7Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T11:58:36.024757Z"},"content_sha256":"629639689981539bb75ba3570462cc45ba27083e0a47aa672b7bc60996c17dfb","schema_version":"1.0","event_id":"sha256:629639689981539bb75ba3570462cc45ba27083e0a47aa672b7bc60996c17dfb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/bundle.json","state_url":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-29T11:58:36Z","links":{"resolver":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL","bundle":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/bundle.json","state":"https://pith.science/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OFHYFHOPBTZKZ6OCKLFTUBGKSL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OFHYFHOPBTZKZ6OCKLFTUBGKSL","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e613205eb7fc444207956f2d552d34a35ef06b68bbfac03f9a4edbceed5fe6de","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-29T18:32:14Z","title_canon_sha256":"4cb4b821ec8f06e080c920e15874735ff5549d8a61a599a8ca75afce2c452644"},"schema_version":"1.0","source":{"id":"2606.00251","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00251","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00251v1","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00251","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"pith_short_12","alias_value":"OFHYFHOPBTZK","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"pith_short_16","alias_value":"OFHYFHOPBTZKZ6OC","created_at":"2026-06-02T01:03:22Z"},{"alias_kind":"pith_short_8","alias_value":"OFHYFHOP","created_at":"2026-06-02T01:03:22Z"}],"graph_snapshots":[{"event_id":"sha256:629639689981539bb75ba3570462cc45ba27083e0a47aa672b7bc60996c17dfb","target":"graph","created_at":"2026-06-02T01:03:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.00251/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Haoyan Yang, Heng Huang, Jiawei Zhou, Reza Shirkavand, Shangqian Gao, Yukai Jin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-29T18:32:14Z","title":"Capability Self-Assessment: Teaching LLMs to Know Their Limits"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00251","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4fe6adcaeedf178c525e0f499215bb8518fbbb5a827b93392e53b78b8db5e9ca","target":"record","created_at":"2026-06-02T01:03:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e613205eb7fc444207956f2d552d34a35ef06b68bbfac03f9a4edbceed5fe6de","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-29T18:32:14Z","title_canon_sha256":"4cb4b821ec8f06e080c920e15874735ff5549d8a61a599a8ca75afce2c452644"},"schema_version":"1.0","source":{"id":"2606.00251","kind":"arxiv","version":1}},"canonical_sha256":"714f829dcf0cf2acf9c252cb3a04ca92e85bd1d6ced8bc641c0cc429ef6ef9e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"714f829dcf0cf2acf9c252cb3a04ca92e85bd1d6ced8bc641c0cc429ef6ef9e0","first_computed_at":"2026-06-02T01:03:22.788067Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:22.788067Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6OP9QJ4hUHRY3JMrjtJROO7IqdjokRZyBHZeBUXP4iTJY8AhlPP9RsIJ2y/QFjvjUOR5ccRuOzrEnNB/kR0fBA==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:22.788431Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00251","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4fe6adcaeedf178c525e0f499215bb8518fbbb5a827b93392e53b78b8db5e9ca","sha256:629639689981539bb75ba3570462cc45ba27083e0a47aa672b7bc60996c17dfb"],"state_sha256":"cce1dd9b42fc009390d3d0560b6b6e1a0c633729a5d5ceac666c7d1fd6b97b8f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cM7vCJM07YQl2buijyyA/N/EZ3MlhVFlPSE1SEqpmMhipfxhXXBbeqgHtm+MG3a/y0R3NSv5B+KRqO14fpOYDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-29T11:58:36.026698Z","bundle_sha256":"8c99a68ba01d7870f9e57a360c28704623f984066e01c87f4d573bb5cbc15efa"}}