{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:64BT7HGN44KRMHU4DWR5EIJD2G","short_pith_number":"pith:64BT7HGN","canonical_record":{"source":{"id":"2403.08819","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-20T04:13:48Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"6befb37658441f40bdc9739ef9ca4bc272526fb8125b95fbe39f8ed9fe9e4c81","abstract_canon_sha256":"0e5600c0522b6202a55e517e655f04d1b1fcd7d9adc7f94dea68c1f938993ff9"},"schema_version":"1.0"},"canonical_sha256":"f7033f9ccde715161e9c1da3d22123d1a39b0cd8fd78cb4a28573919cb81cff1","source":{"kind":"arxiv","id":"2403.08819","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.08819","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"arxiv_version","alias_value":"2403.08819v2","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.08819","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"pith_short_12","alias_value":"64BT7HGN44KR","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"pith_short_16","alias_value":"64BT7HGN44KRMHU4","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"pith_short_8","alias_value":"64BT7HGN","created_at":"2026-07-05T08:37:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:64BT7HGN44KRMHU4DWR5EIJD2G","target":"record","payload":{"canonical_record":{"source":{"id":"2403.08819","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-20T04:13:48Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"6befb37658441f40bdc9739ef9ca4bc272526fb8125b95fbe39f8ed9fe9e4c81","abstract_canon_sha256":"0e5600c0522b6202a55e517e655f04d1b1fcd7d9adc7f94dea68c1f938993ff9"},"schema_version":"1.0"},"canonical_sha256":"f7033f9ccde715161e9c1da3d22123d1a39b0cd8fd78cb4a28573919cb81cff1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:37:18.667283Z","signature_b64":"h6goERq4lVbJi6Ly8D2UE2OkLZHLhqVUCTLg2hIBVFpEWurd6sE1mNrmqd5Tw26vsl/xvJxq3bjKLTlmvL84DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7033f9ccde715161e9c1da3d22123d1a39b0cd8fd78cb4a28573919cb81cff1","last_reissued_at":"2026-07-05T08:37:18.666786Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:37:18.666786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.08819","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":"2026-07-05T08:37:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"X9ZTRnaWOCTOnEEqaLOZO1sCiv5xf7x1RKK85yXnXqMfAjdEgQQosYKqOLJQhq1Lg1VG71vOCUkKKwxc+efRBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:30:05.623588Z"},"content_sha256":"6f43373d04d6b48416e1adf706bebb7b2ad158f52815c1f6629ca7b5eed7b650","schema_version":"1.0","event_id":"sha256:6f43373d04d6b48416e1adf706bebb7b2ad158f52815c1f6629ca7b5eed7b650"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:64BT7HGN44KRMHU4DWR5EIJD2G","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Thermometer: Towards Universal Calibration for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Gregory Wornell, Kristjan Greenewald, Maohao Shen, Prasanna Sattigeri, Soumya Ghosh, Subhro Das","submitted_at":"2024-02-20T04:13:48Z","abstract_excerpt":"We consider the issue of calibration in large language models (LLM). Recent studies have found that common interventions such as instruction tuning often result in poorly calibrated LLMs. Although calibration is well-explored in traditional applications, calibrating LLMs is uniquely challenging. These challenges stem as much from the severe computational requirements of LLMs as from their versatility, which allows them to be applied to diverse tasks. Addressing these challenges, we propose THERMOMETER, a calibration approach tailored to LLMs. THERMOMETER learns an auxiliary model, given data f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.08819","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/2403.08819/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":"2026-07-05T08:37:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r1l6JYOlNDHvXPfijnbnvCVNTNYW9N1hsBtlJ4aeC4VZR/JW91CXc/f6d07+tCPv2fGKRrykBw1xC+DpD6KFCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:30:05.625269Z"},"content_sha256":"96ebe1fb4b8146791af4a0bb736523b72be1975aae542e28d503898008a6609d","schema_version":"1.0","event_id":"sha256:96ebe1fb4b8146791af4a0bb736523b72be1975aae542e28d503898008a6609d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/64BT7HGN44KRMHU4DWR5EIJD2G/bundle.json","state_url":"https://pith.science/pith/64BT7HGN44KRMHU4DWR5EIJD2G/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/64BT7HGN44KRMHU4DWR5EIJD2G/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":"2026-08-05T13:30:05Z","links":{"resolver":"https://pith.science/pith/64BT7HGN44KRMHU4DWR5EIJD2G","bundle":"https://pith.science/pith/64BT7HGN44KRMHU4DWR5EIJD2G/bundle.json","state":"https://pith.science/pith/64BT7HGN44KRMHU4DWR5EIJD2G/state.json","well_known_bundle":"https://pith.science/.well-known/pith/64BT7HGN44KRMHU4DWR5EIJD2G/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:64BT7HGN44KRMHU4DWR5EIJD2G","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":"0e5600c0522b6202a55e517e655f04d1b1fcd7d9adc7f94dea68c1f938993ff9","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-20T04:13:48Z","title_canon_sha256":"6befb37658441f40bdc9739ef9ca4bc272526fb8125b95fbe39f8ed9fe9e4c81"},"schema_version":"1.0","source":{"id":"2403.08819","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.08819","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"arxiv_version","alias_value":"2403.08819v2","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.08819","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"pith_short_12","alias_value":"64BT7HGN44KR","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"pith_short_16","alias_value":"64BT7HGN44KRMHU4","created_at":"2026-07-05T08:37:18Z"},{"alias_kind":"pith_short_8","alias_value":"64BT7HGN","created_at":"2026-07-05T08:37:18Z"}],"graph_snapshots":[{"event_id":"sha256:96ebe1fb4b8146791af4a0bb736523b72be1975aae542e28d503898008a6609d","target":"graph","created_at":"2026-07-05T08:37:18Z","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/2403.08819/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We consider the issue of calibration in large language models (LLM). Recent studies have found that common interventions such as instruction tuning often result in poorly calibrated LLMs. Although calibration is well-explored in traditional applications, calibrating LLMs is uniquely challenging. These challenges stem as much from the severe computational requirements of LLMs as from their versatility, which allows them to be applied to diverse tasks. Addressing these challenges, we propose THERMOMETER, a calibration approach tailored to LLMs. THERMOMETER learns an auxiliary model, given data f","authors_text":"Gregory Wornell, Kristjan Greenewald, Maohao Shen, Prasanna Sattigeri, Soumya Ghosh, Subhro Das","cross_cats":["cs.CL","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-20T04:13:48Z","title":"Thermometer: Towards Universal Calibration for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.08819","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":"sha256:6f43373d04d6b48416e1adf706bebb7b2ad158f52815c1f6629ca7b5eed7b650","target":"record","created_at":"2026-07-05T08:37:18Z","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":"0e5600c0522b6202a55e517e655f04d1b1fcd7d9adc7f94dea68c1f938993ff9","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-20T04:13:48Z","title_canon_sha256":"6befb37658441f40bdc9739ef9ca4bc272526fb8125b95fbe39f8ed9fe9e4c81"},"schema_version":"1.0","source":{"id":"2403.08819","kind":"arxiv","version":2}},"canonical_sha256":"f7033f9ccde715161e9c1da3d22123d1a39b0cd8fd78cb4a28573919cb81cff1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7033f9ccde715161e9c1da3d22123d1a39b0cd8fd78cb4a28573919cb81cff1","first_computed_at":"2026-07-05T08:37:18.666786Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:37:18.666786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"h6goERq4lVbJi6Ly8D2UE2OkLZHLhqVUCTLg2hIBVFpEWurd6sE1mNrmqd5Tw26vsl/xvJxq3bjKLTlmvL84DA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:37:18.667283Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.08819","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6f43373d04d6b48416e1adf706bebb7b2ad158f52815c1f6629ca7b5eed7b650","sha256:96ebe1fb4b8146791af4a0bb736523b72be1975aae542e28d503898008a6609d"],"state_sha256":"a21e3d27ebe35f0024b870e546c1cc05a32c772b536c2d2485e8b0208bd59377"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SmRJLEPEHzB8bC8EJaJCIUYsE6k4j0GwAahL3UCncmiDLkhB3vZZ4k2HFpSKTUygrJJKXTAw2MiEOlloCQCxBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T13:30:05.633726Z","bundle_sha256":"b5bc19adf3a644220723187f7806a2117c67b851f47eebd39edfb061c55bcd62"}}