{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DHPJL57ND2YD4QQJBMSH24PIKJ","short_pith_number":"pith:DHPJL57N","schema_version":"1.0","canonical_sha256":"19de95f7ed1eb03e42090b247d71e852694dfcbcd9799b330fe6454fee347c79","source":{"kind":"arxiv","id":"2410.16509","version":1},"attestation_state":"computed","paper":{"title":"Learning from others' mistakes: Finetuning machine translation models with span-level error annotations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Firas Trabelsi, Hamid Dadkhahi, Jiaming Luo, Lily H. Zhang, Mara Finkelstein, Markus Freitag","submitted_at":"2024-10-21T21:00:47Z","abstract_excerpt":"Despite growing interest in incorporating feedback to improve language models, most efforts focus only on sequence-level annotations. In this work, we explore the potential of utilizing fine-grained span-level annotations from offline datasets to improve model quality. We develop a simple finetuning algorithm, called Training with Annotations (TWA), to directly train machine translation models on such annotated data. TWA utilizes targeted span-level error information while also flexibly learning what to penalize within a span. Moreover, TWA considers the overall trajectory of a sequence when d"},"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":"2410.16509","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-21T21:00:47Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c855d2e89bd87dc7798c8a7e516f007f1ee59044d241ceb94bdac7e182c4c2cf","abstract_canon_sha256":"a69897889539c1483763e593e5ee819e50ce205e79ee4850ebccc7ac6d10be35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:24:02.714396Z","signature_b64":"H5oaHVjShGAVsF70kgB4oQjr0MfUXwp4ULbHAyWGfmvPwsABXnSeYO+ZxJj3tL+OvMBuU0Q4FPoWEi8tuyblAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19de95f7ed1eb03e42090b247d71e852694dfcbcd9799b330fe6454fee347c79","last_reissued_at":"2026-07-05T09:24:02.713938Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:24:02.713938Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning from others' mistakes: Finetuning machine translation models with span-level error annotations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Firas Trabelsi, Hamid Dadkhahi, Jiaming Luo, Lily H. Zhang, Mara Finkelstein, Markus Freitag","submitted_at":"2024-10-21T21:00:47Z","abstract_excerpt":"Despite growing interest in incorporating feedback to improve language models, most efforts focus only on sequence-level annotations. In this work, we explore the potential of utilizing fine-grained span-level annotations from offline datasets to improve model quality. We develop a simple finetuning algorithm, called Training with Annotations (TWA), to directly train machine translation models on such annotated data. TWA utilizes targeted span-level error information while also flexibly learning what to penalize within a span. Moreover, TWA considers the overall trajectory of a sequence when d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.16509","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/2410.16509/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":"2410.16509","created_at":"2026-07-05T09:24:02.714001+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.16509v1","created_at":"2026-07-05T09:24:02.714001+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.16509","created_at":"2026-07-05T09:24:02.714001+00:00"},{"alias_kind":"pith_short_12","alias_value":"DHPJL57ND2YD","created_at":"2026-07-05T09:24:02.714001+00:00"},{"alias_kind":"pith_short_16","alias_value":"DHPJL57ND2YD4QQJ","created_at":"2026-07-05T09:24:02.714001+00:00"},{"alias_kind":"pith_short_8","alias_value":"DHPJL57N","created_at":"2026-07-05T09:24:02.714001+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/DHPJL57ND2YD4QQJBMSH24PIKJ","json":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ.json","graph_json":"https://pith.science/api/pith-number/DHPJL57ND2YD4QQJBMSH24PIKJ/graph.json","events_json":"https://pith.science/api/pith-number/DHPJL57ND2YD4QQJBMSH24PIKJ/events.json","paper":"https://pith.science/paper/DHPJL57N"},"agent_actions":{"view_html":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ","download_json":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ.json","view_paper":"https://pith.science/paper/DHPJL57N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.16509&json=true","fetch_graph":"https://pith.science/api/pith-number/DHPJL57ND2YD4QQJBMSH24PIKJ/graph.json","fetch_events":"https://pith.science/api/pith-number/DHPJL57ND2YD4QQJBMSH24PIKJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/action/storage_attestation","attest_author":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/action/author_attestation","sign_citation":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/action/citation_signature","submit_replication":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/action/replication_record"}},"created_at":"2026-07-05T09:24:02.714001+00:00","updated_at":"2026-07-05T09:24:02.714001+00:00"}