{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QLHBVG2KR6XPVMHMJDOZVZQIRD","short_pith_number":"pith:QLHBVG2K","canonical_record":{"source":{"id":"2604.17487","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-19T15:20:02Z","cross_cats_sorted":[],"title_canon_sha256":"cc88b13115c7162b2b903bd8acccb0c863cdc1e6d540546926d46cbe0305cf88","abstract_canon_sha256":"d4a43b8c3582981ec8e1af52598af73f7ecfec553ed06fa72a8141b263cc4a04"},"schema_version":"1.0"},"canonical_sha256":"82ce1a9b4a8faefab0ec48dd9ae60888e92bd215fe4cb112975947bec3d002c6","source":{"kind":"arxiv","id":"2604.17487","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.17487","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"arxiv_version","alias_value":"2604.17487v2","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.17487","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"pith_short_12","alias_value":"QLHBVG2KR6XP","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"pith_short_16","alias_value":"QLHBVG2KR6XPVMHM","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"pith_short_8","alias_value":"QLHBVG2K","created_at":"2026-05-20T00:04:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QLHBVG2KR6XPVMHMJDOZVZQIRD","target":"record","payload":{"canonical_record":{"source":{"id":"2604.17487","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-19T15:20:02Z","cross_cats_sorted":[],"title_canon_sha256":"cc88b13115c7162b2b903bd8acccb0c863cdc1e6d540546926d46cbe0305cf88","abstract_canon_sha256":"d4a43b8c3582981ec8e1af52598af73f7ecfec553ed06fa72a8141b263cc4a04"},"schema_version":"1.0"},"canonical_sha256":"82ce1a9b4a8faefab0ec48dd9ae60888e92bd215fe4cb112975947bec3d002c6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:31.975024Z","signature_b64":"tj0LSoaPx0/PjP4tEmAcNGc0asBSdzNPh3z6Pi+LH4f/2lwrhWTYO9RKIz0iOIVJWDZiGx0f18Tmgdoc27ClCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"82ce1a9b4a8faefab0ec48dd9ae60888e92bd215fe4cb112975947bec3d002c6","last_reissued_at":"2026-05-20T00:04:31.974207Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:31.974207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.17487","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xoVhLURXdhK9Jj7/ybrv05992CO1kfAR4jEDnrPp3vjU/knaI+2cDvePW48ECxHWwr18RX51lfvafJFyCFlmDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T12:00:28.111330Z"},"content_sha256":"dcb1b836cf4368c7004b801cbe24a6f030f7ecbaf2d32f15ad7dfd16c70317f3","schema_version":"1.0","event_id":"sha256:dcb1b836cf4368c7004b801cbe24a6f030f7ecbaf2d32f15ad7dfd16c70317f3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QLHBVG2KR6XPVMHMJDOZVZQIRD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Answer Only as Precisely as Justified: Calibrated Claim-Level Specificity Control for Agentic Systems","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Compositional selective specificity calibrates each claim to the most specific level supported by evidence.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jason Tansong Dang, Kimberley Yin, Samuel Xu, Samuel Yan, Tianyi Huang","submitted_at":"2026-04-19T15:20:02Z","abstract_excerpt":"Agentic systems often fail not by being entirely wrong, but by being too precise: a response may be generally useful while particular claims exceed what the evidence supports. We study this failure mode as overcommitment control and introduce compositional selective specificity (CSS), a post-generation layer that decomposes an answer into claims, proposes coarser backoffs, and emits each claim at the most specific calibrated level that appears admissible. The method is designed to express uncertainty as a local semantic backoff rather than as a whole-answer refusal. Across a full LongFact run "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across a full LongFact run and HotpotQA pilots, calibrated CSS improves the risk-utility trade-off of fixed drafts. On the full LongFact run, it raises overcommitment-aware utility from 0.846 to 0.913 relative to the no-CSS output while achieving 0.938 specificity retention.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That automated claim decomposition and backoff proposal can reliably identify which parts of a response exceed the evidence without introducing new inaccuracies or losing critical context.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Compositional selective specificity (CSS) improves overcommitment-aware utility from 0.846 to 0.913 on LongFact while retaining 0.938 specificity by calibrating claim-level backoffs in agentic AI responses.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Compositional selective specificity calibrates each claim to the most specific level supported by evidence.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"cb0e364c603d8502968bd820eb6c45a19f5bede72d958c77189b44446f19cef5"},"source":{"id":"2604.17487","kind":"arxiv","version":2},"verdict":{"id":"8cff34ba-4ba2-4094-a6dc-d9eb6b10753b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T05:27:14.306369Z","strongest_claim":"Across a full LongFact run and HotpotQA pilots, calibrated CSS improves the risk-utility trade-off of fixed drafts. On the full LongFact run, it raises overcommitment-aware utility from 0.846 to 0.913 relative to the no-CSS output while achieving 0.938 specificity retention.","one_line_summary":"Compositional selective specificity (CSS) improves overcommitment-aware utility from 0.846 to 0.913 on LongFact while retaining 0.938 specificity by calibrating claim-level backoffs in agentic AI responses.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That automated claim decomposition and backoff proposal can reliably identify which parts of a response exceed the evidence without introducing new inaccuracies or losing critical context.","pith_extraction_headline":"Compositional selective specificity calibrates each claim to the most specific level supported by evidence."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.17487/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":"8cff34ba-4ba2-4094-a6dc-d9eb6b10753b"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/An7uDx7Wo7M8EvZpEgXJ6CTYZHdOBsinPbf/+EemkzEtIZPvnWYZog/MwgJiyFWQu/C2wjgWiY1WuGTyzzhDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T12:00:28.112247Z"},"content_sha256":"ad18a986023b806459f7231c431c2395acf87f88f70e614aab4f6885eaa1d4e2","schema_version":"1.0","event_id":"sha256:ad18a986023b806459f7231c431c2395acf87f88f70e614aab4f6885eaa1d4e2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD/bundle.json","state_url":"https://pith.science/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD/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":"2026-05-25T12:00:28Z","links":{"resolver":"https://pith.science/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD","bundle":"https://pith.science/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD/bundle.json","state":"https://pith.science/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QLHBVG2KR6XPVMHMJDOZVZQIRD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QLHBVG2KR6XPVMHMJDOZVZQIRD","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":"d4a43b8c3582981ec8e1af52598af73f7ecfec553ed06fa72a8141b263cc4a04","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-19T15:20:02Z","title_canon_sha256":"cc88b13115c7162b2b903bd8acccb0c863cdc1e6d540546926d46cbe0305cf88"},"schema_version":"1.0","source":{"id":"2604.17487","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.17487","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"arxiv_version","alias_value":"2604.17487v2","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.17487","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"pith_short_12","alias_value":"QLHBVG2KR6XP","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"pith_short_16","alias_value":"QLHBVG2KR6XPVMHM","created_at":"2026-05-20T00:04:31Z"},{"alias_kind":"pith_short_8","alias_value":"QLHBVG2K","created_at":"2026-05-20T00:04:31Z"}],"graph_snapshots":[{"event_id":"sha256:ad18a986023b806459f7231c431c2395acf87f88f70e614aab4f6885eaa1d4e2","target":"graph","created_at":"2026-05-20T00:04:31Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Across a full LongFact run and HotpotQA pilots, calibrated CSS improves the risk-utility trade-off of fixed drafts. On the full LongFact run, it raises overcommitment-aware utility from 0.846 to 0.913 relative to the no-CSS output while achieving 0.938 specificity retention."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That automated claim decomposition and backoff proposal can reliably identify which parts of a response exceed the evidence without introducing new inaccuracies or losing critical context."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Compositional selective specificity (CSS) improves overcommitment-aware utility from 0.846 to 0.913 on LongFact while retaining 0.938 specificity by calibrating claim-level backoffs in agentic AI responses."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Compositional selective specificity calibrates each claim to the most specific level supported by evidence."}],"snapshot_sha256":"cb0e364c603d8502968bd820eb6c45a19f5bede72d958c77189b44446f19cef5"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.17487/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Agentic systems often fail not by being entirely wrong, but by being too precise: a response may be generally useful while particular claims exceed what the evidence supports. We study this failure mode as overcommitment control and introduce compositional selective specificity (CSS), a post-generation layer that decomposes an answer into claims, proposes coarser backoffs, and emits each claim at the most specific calibrated level that appears admissible. The method is designed to express uncertainty as a local semantic backoff rather than as a whole-answer refusal. Across a full LongFact run ","authors_text":"Jason Tansong Dang, Kimberley Yin, Samuel Xu, Samuel Yan, Tianyi Huang","cross_cats":[],"headline":"Compositional selective specificity calibrates each claim to the most specific level supported by evidence.","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-19T15:20:02Z","title":"Answer Only as Precisely as Justified: Calibrated Claim-Level Specificity Control for Agentic Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.17487","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T05:27:14.306369Z","id":"8cff34ba-4ba2-4094-a6dc-d9eb6b10753b","model_set":{"reader":"grok-4.3"},"one_line_summary":"Compositional selective specificity (CSS) improves overcommitment-aware utility from 0.846 to 0.913 on LongFact while retaining 0.938 specificity by calibrating claim-level backoffs in agentic AI responses.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Compositional selective specificity calibrates each claim to the most specific level supported by evidence.","strongest_claim":"Across a full LongFact run and HotpotQA pilots, calibrated CSS improves the risk-utility trade-off of fixed drafts. On the full LongFact run, it raises overcommitment-aware utility from 0.846 to 0.913 relative to the no-CSS output while achieving 0.938 specificity retention.","weakest_assumption":"That automated claim decomposition and backoff proposal can reliably identify which parts of a response exceed the evidence without introducing new inaccuracies or losing critical context."}},"verdict_id":"8cff34ba-4ba2-4094-a6dc-d9eb6b10753b"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:dcb1b836cf4368c7004b801cbe24a6f030f7ecbaf2d32f15ad7dfd16c70317f3","target":"record","created_at":"2026-05-20T00:04:31Z","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":"d4a43b8c3582981ec8e1af52598af73f7ecfec553ed06fa72a8141b263cc4a04","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-19T15:20:02Z","title_canon_sha256":"cc88b13115c7162b2b903bd8acccb0c863cdc1e6d540546926d46cbe0305cf88"},"schema_version":"1.0","source":{"id":"2604.17487","kind":"arxiv","version":2}},"canonical_sha256":"82ce1a9b4a8faefab0ec48dd9ae60888e92bd215fe4cb112975947bec3d002c6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"82ce1a9b4a8faefab0ec48dd9ae60888e92bd215fe4cb112975947bec3d002c6","first_computed_at":"2026-05-20T00:04:31.974207Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:31.974207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tj0LSoaPx0/PjP4tEmAcNGc0asBSdzNPh3z6Pi+LH4f/2lwrhWTYO9RKIz0iOIVJWDZiGx0f18Tmgdoc27ClCA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:31.975024Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.17487","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dcb1b836cf4368c7004b801cbe24a6f030f7ecbaf2d32f15ad7dfd16c70317f3","sha256:ad18a986023b806459f7231c431c2395acf87f88f70e614aab4f6885eaa1d4e2"],"state_sha256":"60e5b22d6c76c334ea761eae2840c82d350de364d797ccf9fb68ce437fbf6258"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QSv67KNaSDmcUaeLSWnUMtRDKDz0hAfdppie/f3xn3uXJVRqQhAS3fsnUuap0VlqZTWI7PjPloxlQqN6D4NOCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T12:00:28.116560Z","bundle_sha256":"12756718b68672d00c1d5e31abf4352837752671e1222caf4213d8f861db832e"}}