{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:P73THSSMQWE7STPSGFGMWBDO2T","short_pith_number":"pith:P73THSSM","schema_version":"1.0","canonical_sha256":"7ff733ca4c8589f94df2314ccb046ed4c1edc77be751c30ef2fbabbb36592faf","source":{"kind":"arxiv","id":"2501.13927","version":1},"attestation_state":"computed","paper":{"title":"CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Guofeng Cui, Pichao Wang, Vimal Bhat, Yang Liu, Zemian Ke, Zhu Liu","submitted_at":"2025-01-23T18:59:47Z","abstract_excerpt":"Large language models (LLMs) have shown great potential in natural language processing tasks, but their application to machine translation (MT) remains challenging due to pretraining on English-centric data and the complexity of reinforcement learning from human feedback (RLHF). Direct Preference Optimization (DPO) has emerged as a simpler and more efficient alternative, but its performance depends heavily on the quality of preference data. To address this, we propose Confidence-Reward driven Preference Optimization (CRPO), a novel method that combines reward scores with model confidence to im"},"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":"2501.13927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-23T18:59:47Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"b3b8262ebb6a09d842ca03eb3e0f342544467cd9f6061e61328448481498040b","abstract_canon_sha256":"1d33c6a50ee824186c3c62a77ffda3aa6705a35f4ab0d951adf0950fbee39dec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:36.100171Z","signature_b64":"k0WV2xUy9aQ7EIQ+arK/om2rMofhWN+hL37UE4OYZPctr5Mv/fToc0AerBgmwtilg20Miung4P3BJmFwDvI8Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ff733ca4c8589f94df2314ccb046ed4c1edc77be751c30ef2fbabbb36592faf","last_reissued_at":"2026-07-05T10:04:36.099826Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:36.099826Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Guofeng Cui, Pichao Wang, Vimal Bhat, Yang Liu, Zemian Ke, Zhu Liu","submitted_at":"2025-01-23T18:59:47Z","abstract_excerpt":"Large language models (LLMs) have shown great potential in natural language processing tasks, but their application to machine translation (MT) remains challenging due to pretraining on English-centric data and the complexity of reinforcement learning from human feedback (RLHF). Direct Preference Optimization (DPO) has emerged as a simpler and more efficient alternative, but its performance depends heavily on the quality of preference data. To address this, we propose Confidence-Reward driven Preference Optimization (CRPO), a novel method that combines reward scores with model confidence to im"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13927","kind":"arxiv","version":1},"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/2501.13927/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":"2501.13927","created_at":"2026-07-05T10:04:36.099885+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.13927v1","created_at":"2026-07-05T10:04:36.099885+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13927","created_at":"2026-07-05T10:04:36.099885+00:00"},{"alias_kind":"pith_short_12","alias_value":"P73THSSMQWE7","created_at":"2026-07-05T10:04:36.099885+00:00"},{"alias_kind":"pith_short_16","alias_value":"P73THSSMQWE7STPS","created_at":"2026-07-05T10:04:36.099885+00:00"},{"alias_kind":"pith_short_8","alias_value":"P73THSSM","created_at":"2026-07-05T10:04:36.099885+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T","json":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T.json","graph_json":"https://pith.science/api/pith-number/P73THSSMQWE7STPSGFGMWBDO2T/graph.json","events_json":"https://pith.science/api/pith-number/P73THSSMQWE7STPSGFGMWBDO2T/events.json","paper":"https://pith.science/paper/P73THSSM"},"agent_actions":{"view_html":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T","download_json":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T.json","view_paper":"https://pith.science/paper/P73THSSM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.13927&json=true","fetch_graph":"https://pith.science/api/pith-number/P73THSSMQWE7STPSGFGMWBDO2T/graph.json","fetch_events":"https://pith.science/api/pith-number/P73THSSMQWE7STPSGFGMWBDO2T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T/action/storage_attestation","attest_author":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T/action/author_attestation","sign_citation":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T/action/citation_signature","submit_replication":"https://pith.science/pith/P73THSSMQWE7STPSGFGMWBDO2T/action/replication_record"}},"created_at":"2026-07-05T10:04:36.099885+00:00","updated_at":"2026-07-05T10:04:36.099885+00:00"}