{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OG4JUEGHTFTETZ36BIE2VP7RMP","short_pith_number":"pith:OG4JUEGH","schema_version":"1.0","canonical_sha256":"71b89a10c7996649e77e0a09aabff163f4107e4120a9020d0188319d11cc4c8f","source":{"kind":"arxiv","id":"2502.12601","version":3},"attestation_state":"computed","paper":{"title":"COPU: Conformal Prediction for Uncertainty Quantification in Natural Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hanjie Chen, Lu Cheng, Sean Wang, YiCheng Jiang, Yuxin Tang","submitted_at":"2025-02-18T07:25:12Z","abstract_excerpt":"Uncertainty Quantification (UQ) for Natural Language Generation (NLG) is crucial for assessing the performance of Large Language Models (LLMs), as it reveals confidence in predictions, identifies failure modes, and gauges output reliability. Conformal Prediction (CP), a model-agnostic method that generates prediction sets with a specified error rate, has been adopted for UQ in classification tasks, where the size of the prediction set indicates the model's uncertainty. However, when adapting CP to NLG, the sampling-based method for generating candidate outputs cannot guarantee the inclusion of"},"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":"2502.12601","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T07:25:12Z","cross_cats_sorted":[],"title_canon_sha256":"46590fbfb314c1cede8d567bc57efb547a89f77d075770f0243e6555fba8befe","abstract_canon_sha256":"79ba31b70d9ff9a8f20109c9ce89b9d0017a3aad02f82f40ce3493666776b18f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:00.445299Z","signature_b64":"VahTlNR1c/rpfFGp95x2tKkTcydIPMZZgkA0BiKTqgP8IVMnJtZMprdt9e3Walg/6gP5JqkEGIuebrsdVA6UCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71b89a10c7996649e77e0a09aabff163f4107e4120a9020d0188319d11cc4c8f","last_reissued_at":"2026-07-05T10:46:00.444798Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:00.444798Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"COPU: Conformal Prediction for Uncertainty Quantification in Natural Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hanjie Chen, Lu Cheng, Sean Wang, YiCheng Jiang, Yuxin Tang","submitted_at":"2025-02-18T07:25:12Z","abstract_excerpt":"Uncertainty Quantification (UQ) for Natural Language Generation (NLG) is crucial for assessing the performance of Large Language Models (LLMs), as it reveals confidence in predictions, identifies failure modes, and gauges output reliability. Conformal Prediction (CP), a model-agnostic method that generates prediction sets with a specified error rate, has been adopted for UQ in classification tasks, where the size of the prediction set indicates the model's uncertainty. However, when adapting CP to NLG, the sampling-based method for generating candidate outputs cannot guarantee the inclusion of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12601","kind":"arxiv","version":3},"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/2502.12601/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":"2502.12601","created_at":"2026-07-05T10:46:00.444859+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.12601v3","created_at":"2026-07-05T10:46:00.444859+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12601","created_at":"2026-07-05T10:46:00.444859+00:00"},{"alias_kind":"pith_short_12","alias_value":"OG4JUEGHTFTE","created_at":"2026-07-05T10:46:00.444859+00:00"},{"alias_kind":"pith_short_16","alias_value":"OG4JUEGHTFTETZ36","created_at":"2026-07-05T10:46:00.444859+00:00"},{"alias_kind":"pith_short_8","alias_value":"OG4JUEGH","created_at":"2026-07-05T10:46:00.444859+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02911","citing_title":"The Ghost Annotator: a Framework to Explore Human Label Variation in Content Moderation through Conformal Prediction","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01850","citing_title":"Does Compression Preserve Uncertainty? A Unified Benchmark for Quantized and Sparse LLMs via Conformal Prediction","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP","json":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP.json","graph_json":"https://pith.science/api/pith-number/OG4JUEGHTFTETZ36BIE2VP7RMP/graph.json","events_json":"https://pith.science/api/pith-number/OG4JUEGHTFTETZ36BIE2VP7RMP/events.json","paper":"https://pith.science/paper/OG4JUEGH"},"agent_actions":{"view_html":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP","download_json":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP.json","view_paper":"https://pith.science/paper/OG4JUEGH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.12601&json=true","fetch_graph":"https://pith.science/api/pith-number/OG4JUEGHTFTETZ36BIE2VP7RMP/graph.json","fetch_events":"https://pith.science/api/pith-number/OG4JUEGHTFTETZ36BIE2VP7RMP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP/action/storage_attestation","attest_author":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP/action/author_attestation","sign_citation":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP/action/citation_signature","submit_replication":"https://pith.science/pith/OG4JUEGHTFTETZ36BIE2VP7RMP/action/replication_record"}},"created_at":"2026-07-05T10:46:00.444859+00:00","updated_at":"2026-07-05T10:46:00.444859+00:00"}