{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HQGJW3HKDWUR33JR3DEGELUJL6","short_pith_number":"pith:HQGJW3HK","schema_version":"1.0","canonical_sha256":"3c0c9b6cea1da91ded31d8c8622e895f94941733c368f7c18e8bf7ea85a3263e","source":{"kind":"arxiv","id":"2605.28070","version":1},"attestation_state":"computed","paper":{"title":"Bridging the Detection-to-Abstention Gap in Reasoning Models under Insufficient Information","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chunxiao Guo, Hansong Xiao, Jiaxu Li, Jinjie Gu, Pei Wei, Renjie Gu, Yefei Chen, Yihao Wang, Yixin Cao, Yuan Wang, Yun Yue","submitted_at":"2026-05-27T07:28:25Z","abstract_excerpt":"We highlight a failure mode of large reasoning models on questions with insufficient information: models may recognize that a problem is under-specified, yet still continue reasoning and produce unsupported final answers instead of abstaining. We formalize this mismatch as the detection-to-abstention gap, where detected insufficiency fails to translate into final abstention. This gap is especially concerning in high-risk domains such as medical AI, where answers based on incomplete evidence can be more harmful than refusal. To close this gap, we propose Judge-Then-Solve (JTS), a trajectory-lev"},"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":"2605.28070","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T07:28:25Z","cross_cats_sorted":[],"title_canon_sha256":"1f34f53431a44de9c303cf278437b0641222cd2f325bb46ecaa1cd1d5ede9b3e","abstract_canon_sha256":"db579a61acecfbcceecef345797aeb00bcca463ecc700a9c403ee07bb41af30d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:57.700618Z","signature_b64":"gCb1WdD8eIipT/b/xtP/BbDbHclVFMtJBXDQ7/ZFWIM/0P7iiTwagmMCtqWb9nnrmE0gPc44dw0fSaTmeSz8Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c0c9b6cea1da91ded31d8c8622e895f94941733c368f7c18e8bf7ea85a3263e","last_reissued_at":"2026-05-28T01:04:57.700171Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:57.700171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bridging the Detection-to-Abstention Gap in Reasoning Models under Insufficient Information","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chunxiao Guo, Hansong Xiao, Jiaxu Li, Jinjie Gu, Pei Wei, Renjie Gu, Yefei Chen, Yihao Wang, Yixin Cao, Yuan Wang, Yun Yue","submitted_at":"2026-05-27T07:28:25Z","abstract_excerpt":"We highlight a failure mode of large reasoning models on questions with insufficient information: models may recognize that a problem is under-specified, yet still continue reasoning and produce unsupported final answers instead of abstaining. We formalize this mismatch as the detection-to-abstention gap, where detected insufficiency fails to translate into final abstention. This gap is especially concerning in high-risk domains such as medical AI, where answers based on incomplete evidence can be more harmful than refusal. To close this gap, we propose Judge-Then-Solve (JTS), a trajectory-lev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28070","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/2605.28070/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":"2605.28070","created_at":"2026-05-28T01:04:57.700239+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28070v1","created_at":"2026-05-28T01:04:57.700239+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28070","created_at":"2026-05-28T01:04:57.700239+00:00"},{"alias_kind":"pith_short_12","alias_value":"HQGJW3HKDWUR","created_at":"2026-05-28T01:04:57.700239+00:00"},{"alias_kind":"pith_short_16","alias_value":"HQGJW3HKDWUR33JR","created_at":"2026-05-28T01:04:57.700239+00:00"},{"alias_kind":"pith_short_8","alias_value":"HQGJW3HK","created_at":"2026-05-28T01:04:57.700239+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/HQGJW3HKDWUR33JR3DEGELUJL6","json":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6.json","graph_json":"https://pith.science/api/pith-number/HQGJW3HKDWUR33JR3DEGELUJL6/graph.json","events_json":"https://pith.science/api/pith-number/HQGJW3HKDWUR33JR3DEGELUJL6/events.json","paper":"https://pith.science/paper/HQGJW3HK"},"agent_actions":{"view_html":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6","download_json":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6.json","view_paper":"https://pith.science/paper/HQGJW3HK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28070&json=true","fetch_graph":"https://pith.science/api/pith-number/HQGJW3HKDWUR33JR3DEGELUJL6/graph.json","fetch_events":"https://pith.science/api/pith-number/HQGJW3HKDWUR33JR3DEGELUJL6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6/action/storage_attestation","attest_author":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6/action/author_attestation","sign_citation":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6/action/citation_signature","submit_replication":"https://pith.science/pith/HQGJW3HKDWUR33JR3DEGELUJL6/action/replication_record"}},"created_at":"2026-05-28T01:04:57.700239+00:00","updated_at":"2026-05-28T01:04:57.700239+00:00"}