{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KCX2MIURCVV7CGVL4C4PRRKM5U","short_pith_number":"pith:KCX2MIUR","canonical_record":{"source":{"id":"2409.00352","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-31T05:12:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"dd674449c688e98949c7ff5e1fca8531b25710c938e44e5de53379f24d8a687e","abstract_canon_sha256":"bb6311877a9f3114fce566035a05f13d5a066edec33f2c404dd1fd2b0120e415"},"schema_version":"1.0"},"canonical_sha256":"50afa62291156bf11aabe0b8f8c54ced3010310c6664dbfb88a5fded1dc995bd","source":{"kind":"arxiv","id":"2409.00352","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.00352","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"arxiv_version","alias_value":"2409.00352v2","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.00352","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"pith_short_12","alias_value":"KCX2MIURCVV7","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"pith_short_16","alias_value":"KCX2MIURCVV7CGVL","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"pith_short_8","alias_value":"KCX2MIUR","created_at":"2026-07-05T10:11:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KCX2MIURCVV7CGVL4C4PRRKM5U","target":"record","payload":{"canonical_record":{"source":{"id":"2409.00352","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-31T05:12:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"dd674449c688e98949c7ff5e1fca8531b25710c938e44e5de53379f24d8a687e","abstract_canon_sha256":"bb6311877a9f3114fce566035a05f13d5a066edec33f2c404dd1fd2b0120e415"},"schema_version":"1.0"},"canonical_sha256":"50afa62291156bf11aabe0b8f8c54ced3010310c6664dbfb88a5fded1dc995bd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:11:23.585427Z","signature_b64":"SKxTsqS+Hs7OHikXoQyTZ0efNGvHRlVF6jS4jYh6jXmcT+rqjflk1KWtTHdhhUyuSHx/zwCS0MJdNjEtWWl5Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50afa62291156bf11aabe0b8f8c54ced3010310c6664dbfb88a5fded1dc995bd","last_reissued_at":"2026-07-05T10:11:23.584920Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:11:23.584920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.00352","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":"2026-07-05T10:11:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pxtLJEEzlQZM+VpgVtWYkRx3kf45qFUIqyz6uzCpsr3KqP9aPwc00eGXLww8Vgj/o3zG4Nv8EV8GjJazekqQCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T21:03:19.053135Z"},"content_sha256":"3717c65bb4e1260899eec95ebd7f5cb42e58975352d07cfc1cd2ff2a242ee4b9","schema_version":"1.0","event_id":"sha256:3717c65bb4e1260899eec95ebd7f5cb42e58975352d07cfc1cd2ff2a242ee4b9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KCX2MIURCVV7CGVL4C4PRRKM5U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Does Alignment Tuning Really Break LLMs' Internal Confidence?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongseok Oh, Wonseok Hwang","submitted_at":"2024-08-31T05:12:36Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable progress, but their real-world application necessitates reliable calibration. This study conducts a comprehensive analysis of calibration degradation of LLMs across four dimensions: models, calibration metrics, tasks, and confidence extraction methods. Initial analysis showed that the relationship between alignment and calibration is not always a trade-off, but under stricter analysis conditions, we found the alignment process consistently harms calibration. This highlights the need for (1) a careful approach when measuring model confidences a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.00352","kind":"arxiv","version":2},"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/2409.00352/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":"2026-07-05T10:11:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OpnBrjRIaPLpYI+KN2CycDE9AM7QF2PbLt4+UI9TGdZHfEiQ9/uGHlY9JlZyU9AKaiOFlUOkr8kJ65wqODkuCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T21:03:19.053651Z"},"content_sha256":"a9a7181749be7840d95191705f9d75afeb15a5164655dc54a154e32e4c39f657","schema_version":"1.0","event_id":"sha256:a9a7181749be7840d95191705f9d75afeb15a5164655dc54a154e32e4c39f657"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KCX2MIURCVV7CGVL4C4PRRKM5U/bundle.json","state_url":"https://pith.science/pith/KCX2MIURCVV7CGVL4C4PRRKM5U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KCX2MIURCVV7CGVL4C4PRRKM5U/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":"2026-08-10T21:03:19Z","links":{"resolver":"https://pith.science/pith/KCX2MIURCVV7CGVL4C4PRRKM5U","bundle":"https://pith.science/pith/KCX2MIURCVV7CGVL4C4PRRKM5U/bundle.json","state":"https://pith.science/pith/KCX2MIURCVV7CGVL4C4PRRKM5U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KCX2MIURCVV7CGVL4C4PRRKM5U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KCX2MIURCVV7CGVL4C4PRRKM5U","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":"bb6311877a9f3114fce566035a05f13d5a066edec33f2c404dd1fd2b0120e415","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-31T05:12:36Z","title_canon_sha256":"dd674449c688e98949c7ff5e1fca8531b25710c938e44e5de53379f24d8a687e"},"schema_version":"1.0","source":{"id":"2409.00352","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.00352","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"arxiv_version","alias_value":"2409.00352v2","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.00352","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"pith_short_12","alias_value":"KCX2MIURCVV7","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"pith_short_16","alias_value":"KCX2MIURCVV7CGVL","created_at":"2026-07-05T10:11:23Z"},{"alias_kind":"pith_short_8","alias_value":"KCX2MIUR","created_at":"2026-07-05T10:11:23Z"}],"graph_snapshots":[{"event_id":"sha256:a9a7181749be7840d95191705f9d75afeb15a5164655dc54a154e32e4c39f657","target":"graph","created_at":"2026-07-05T10:11:23Z","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/2409.00352/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have shown remarkable progress, but their real-world application necessitates reliable calibration. This study conducts a comprehensive analysis of calibration degradation of LLMs across four dimensions: models, calibration metrics, tasks, and confidence extraction methods. Initial analysis showed that the relationship between alignment and calibration is not always a trade-off, but under stricter analysis conditions, we found the alignment process consistently harms calibration. This highlights the need for (1) a careful approach when measuring model confidences a","authors_text":"Hongseok Oh, Wonseok Hwang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-31T05:12:36Z","title":"Does Alignment Tuning Really Break LLMs' Internal Confidence?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.00352","kind":"arxiv","version":2},"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":"sha256:3717c65bb4e1260899eec95ebd7f5cb42e58975352d07cfc1cd2ff2a242ee4b9","target":"record","created_at":"2026-07-05T10:11:23Z","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":"bb6311877a9f3114fce566035a05f13d5a066edec33f2c404dd1fd2b0120e415","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-31T05:12:36Z","title_canon_sha256":"dd674449c688e98949c7ff5e1fca8531b25710c938e44e5de53379f24d8a687e"},"schema_version":"1.0","source":{"id":"2409.00352","kind":"arxiv","version":2}},"canonical_sha256":"50afa62291156bf11aabe0b8f8c54ced3010310c6664dbfb88a5fded1dc995bd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"50afa62291156bf11aabe0b8f8c54ced3010310c6664dbfb88a5fded1dc995bd","first_computed_at":"2026-07-05T10:11:23.584920Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:11:23.584920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SKxTsqS+Hs7OHikXoQyTZ0efNGvHRlVF6jS4jYh6jXmcT+rqjflk1KWtTHdhhUyuSHx/zwCS0MJdNjEtWWl5Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:11:23.585427Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.00352","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3717c65bb4e1260899eec95ebd7f5cb42e58975352d07cfc1cd2ff2a242ee4b9","sha256:a9a7181749be7840d95191705f9d75afeb15a5164655dc54a154e32e4c39f657"],"state_sha256":"479075138f97392a66a653b750e025c1544fc6730516103cec8db063d105b9d0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qrY+EEKs3c6m/U4PJOEZknvHx2Ob5G9D2re6MuulW252tMxVqzaFhC1p8HuTDERxErqUpt7mr6rlzpPrkOl6Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T21:03:19.058006Z","bundle_sha256":"71d59c2665d5902ebdfbe466d854524caf14f044554bc620d9dc1d226b2f10fe"}}