{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:RT4RGK54JKRLM3MHR4ECR4XBQG","short_pith_number":"pith:RT4RGK54","schema_version":"1.0","canonical_sha256":"8cf9132bbc4aa2b66d878f0828f2e181af8251bf225b95fae39411cd6d7c8364","source":{"kind":"arxiv","id":"2205.12507","version":2},"attestation_state":"computed","paper":{"title":"Re-Examining Calibration: The Case of Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenglei Si, Chen Zhao, Jordan Boyd-Graber, Sewon Min","submitted_at":"2022-05-25T05:49:56Z","abstract_excerpt":"For users to trust model predictions, they need to understand model outputs, particularly their confidence - calibration aims to adjust (calibrate) models' confidence to match expected accuracy. We argue that the traditional calibration evaluation does not promote effective calibrations: for example, it can encourage always assigning a mediocre confidence score to all predictions, which does not help users distinguish correct predictions from wrong ones. Building on those observations, we propose a new calibration metric, MacroCE, that better captures whether the model assigns low confidence t"},"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":"2205.12507","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-25T05:49:56Z","cross_cats_sorted":[],"title_canon_sha256":"44e42cc6dc89e5b3101457fd9d88cc631c6e3e14e49a3b7c58a0944a43cff459","abstract_canon_sha256":"1d0985e04773c4c26d47598e50aff894d3f2af337a3d3963501fb49760a2a2d4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:09:18.577927Z","signature_b64":"VEjFiAhE4sORVwdUifgxjfQgLW7ihDRdU7XNDMlL+uSNqNsVr02cBpoI+AugbE1skXiK6h5Rvh8H3IJzL+NTCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8cf9132bbc4aa2b66d878f0828f2e181af8251bf225b95fae39411cd6d7c8364","last_reissued_at":"2026-07-05T05:09:18.577520Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:09:18.577520Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Re-Examining Calibration: The Case of Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenglei Si, Chen Zhao, Jordan Boyd-Graber, Sewon Min","submitted_at":"2022-05-25T05:49:56Z","abstract_excerpt":"For users to trust model predictions, they need to understand model outputs, particularly their confidence - calibration aims to adjust (calibrate) models' confidence to match expected accuracy. We argue that the traditional calibration evaluation does not promote effective calibrations: for example, it can encourage always assigning a mediocre confidence score to all predictions, which does not help users distinguish correct predictions from wrong ones. Building on those observations, we propose a new calibration metric, MacroCE, that better captures whether the model assigns low confidence t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.12507","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/2205.12507/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":"2205.12507","created_at":"2026-07-05T05:09:18.577574+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.12507v2","created_at":"2026-07-05T05:09:18.577574+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.12507","created_at":"2026-07-05T05:09:18.577574+00:00"},{"alias_kind":"pith_short_12","alias_value":"RT4RGK54JKRL","created_at":"2026-07-05T05:09:18.577574+00:00"},{"alias_kind":"pith_short_16","alias_value":"RT4RGK54JKRLM3MH","created_at":"2026-07-05T05:09:18.577574+00:00"},{"alias_kind":"pith_short_8","alias_value":"RT4RGK54","created_at":"2026-07-05T05:09:18.577574+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19950","citing_title":"Confidence Calibration for Multimodal LLMs: An Empirical Study through Medical VQA","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09106","citing_title":"Fin-Bias: Comprehensive Evaluation for LLM Decision-Making under human bias in Finance Domain","ref_index":71,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG","json":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG.json","graph_json":"https://pith.science/api/pith-number/RT4RGK54JKRLM3MHR4ECR4XBQG/graph.json","events_json":"https://pith.science/api/pith-number/RT4RGK54JKRLM3MHR4ECR4XBQG/events.json","paper":"https://pith.science/paper/RT4RGK54"},"agent_actions":{"view_html":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG","download_json":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG.json","view_paper":"https://pith.science/paper/RT4RGK54","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.12507&json=true","fetch_graph":"https://pith.science/api/pith-number/RT4RGK54JKRLM3MHR4ECR4XBQG/graph.json","fetch_events":"https://pith.science/api/pith-number/RT4RGK54JKRLM3MHR4ECR4XBQG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG/action/storage_attestation","attest_author":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG/action/author_attestation","sign_citation":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG/action/citation_signature","submit_replication":"https://pith.science/pith/RT4RGK54JKRLM3MHR4ECR4XBQG/action/replication_record"}},"created_at":"2026-07-05T05:09:18.577574+00:00","updated_at":"2026-07-05T05:09:18.577574+00:00"}