{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:K7G5MXICRSDS5BYPCYI44KXSJT","short_pith_number":"pith:K7G5MXIC","schema_version":"1.0","canonical_sha256":"57cdd65d028c872e870f1611ce2af24cf8ea6f9cfd1aee6a0ccf19b1f641a523","source":{"kind":"arxiv","id":"2312.04021","version":4},"attestation_state":"computed","paper":{"title":"A Study on the Calibration of In-context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Dean Foster, Dhruv Madeka, Eric Xing, Hanlin Zhang, Himabindu Lakkaraju, Sham Kakade, Yaodong Yu, Yi-Fan Zhang","submitted_at":"2023-12-07T03:37:39Z","abstract_excerpt":"Accurate uncertainty quantification is crucial for the safe deployment of machine learning models, and prior research has demonstrated improvements in the calibration of modern language models (LMs). We study in-context learning (ICL), a prevalent method for adapting static LMs through tailored prompts, and examine the balance between performance and calibration across a broad spectrum of natural language understanding and reasoning tasks. Through comprehensive experiments, we observe that, with an increasing number of ICL examples, models initially exhibit increased miscalibration before achi"},"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":"2312.04021","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-12-07T03:37:39Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"58e7b51fb683283f872123f6eced14db670631263453ba7ec5599825666f04c2","abstract_canon_sha256":"a9eed09ab741d2d32765add25e49e5b09d49feb34732c8a4b36d326e3df68ee3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:35.626472Z","signature_b64":"fOEJaCPrkv6pnZGqVNAmHjpl/xomNjE0VVQbHEFKgUxsCTrNQr2uUp4oDSc1EF8FXNsAgHh7ghoNE4PUyJAXCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"57cdd65d028c872e870f1611ce2af24cf8ea6f9cfd1aee6a0ccf19b1f641a523","last_reissued_at":"2026-07-05T08:01:35.625911Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:35.625911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Study on the Calibration of In-context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Dean Foster, Dhruv Madeka, Eric Xing, Hanlin Zhang, Himabindu Lakkaraju, Sham Kakade, Yaodong Yu, Yi-Fan Zhang","submitted_at":"2023-12-07T03:37:39Z","abstract_excerpt":"Accurate uncertainty quantification is crucial for the safe deployment of machine learning models, and prior research has demonstrated improvements in the calibration of modern language models (LMs). We study in-context learning (ICL), a prevalent method for adapting static LMs through tailored prompts, and examine the balance between performance and calibration across a broad spectrum of natural language understanding and reasoning tasks. Through comprehensive experiments, we observe that, with an increasing number of ICL examples, models initially exhibit increased miscalibration before achi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.04021","kind":"arxiv","version":4},"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/2312.04021/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":"2312.04021","created_at":"2026-07-05T08:01:35.625975+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.04021v4","created_at":"2026-07-05T08:01:35.625975+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.04021","created_at":"2026-07-05T08:01:35.625975+00:00"},{"alias_kind":"pith_short_12","alias_value":"K7G5MXICRSDS","created_at":"2026-07-05T08:01:35.625975+00:00"},{"alias_kind":"pith_short_16","alias_value":"K7G5MXICRSDS5BYP","created_at":"2026-07-05T08:01:35.625975+00:00"},{"alias_kind":"pith_short_8","alias_value":"K7G5MXIC","created_at":"2026-07-05T08:01:35.625975+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17200","citing_title":"Calibrating Model-Based Evaluation Metrics for Summarization","ref_index":72,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT","json":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT.json","graph_json":"https://pith.science/api/pith-number/K7G5MXICRSDS5BYPCYI44KXSJT/graph.json","events_json":"https://pith.science/api/pith-number/K7G5MXICRSDS5BYPCYI44KXSJT/events.json","paper":"https://pith.science/paper/K7G5MXIC"},"agent_actions":{"view_html":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT","download_json":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT.json","view_paper":"https://pith.science/paper/K7G5MXIC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.04021&json=true","fetch_graph":"https://pith.science/api/pith-number/K7G5MXICRSDS5BYPCYI44KXSJT/graph.json","fetch_events":"https://pith.science/api/pith-number/K7G5MXICRSDS5BYPCYI44KXSJT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT/action/storage_attestation","attest_author":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT/action/author_attestation","sign_citation":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT/action/citation_signature","submit_replication":"https://pith.science/pith/K7G5MXICRSDS5BYPCYI44KXSJT/action/replication_record"}},"created_at":"2026-07-05T08:01:35.625975+00:00","updated_at":"2026-07-05T08:01:35.625975+00:00"}