{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DHPJL57ND2YD4QQJBMSH24PIKJ","short_pith_number":"pith:DHPJL57N","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"},"canonical_sha256":"19de95f7ed1eb03e42090b247d71e852694dfcbcd9799b330fe6454fee347c79","source":{"kind":"arxiv","id":"2410.16509","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.16509","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"arxiv_version","alias_value":"2410.16509v1","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.16509","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"pith_short_12","alias_value":"DHPJL57ND2YD","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"pith_short_16","alias_value":"DHPJL57ND2YD4QQJ","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"pith_short_8","alias_value":"DHPJL57N","created_at":"2026-07-05T09:24:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DHPJL57ND2YD4QQJBMSH24PIKJ","target":"record","payload":{"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"},"canonical_sha256":"19de95f7ed1eb03e42090b247d71e852694dfcbcd9799b330fe6454fee347c79","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"},"source_kind":"arxiv","source_id":"2410.16509","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:24:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WpfmHRmIJNYtufGqUFXq9k9sz3urVYwC1z68K0ocLa05aG8olT4Sy0yVQWESBdDkSFZYvYz4k9044Jtb3WmzCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T07:37:51.333705Z"},"content_sha256":"b85e2516e1e33d95da420977a5734081ea31a056dd8c8c5d9ab003bc63ca03c6","schema_version":"1.0","event_id":"sha256:b85e2516e1e33d95da420977a5734081ea31a056dd8c8c5d9ab003bc63ca03c6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DHPJL57ND2YD4QQJBMSH24PIKJ","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:24:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LlP11nBZacP5woK8r7P0kyafdl2TOXYx1VE7gVZrThmZPfH9X5xo3jRiYlLPQMw8aHkn8ChAZiqTl8Cip0R3Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T07:37:51.334272Z"},"content_sha256":"06a7f7acd633fa921b8c6b0ae85b48bb69c5d66d611bc2dcaab18e42d07a888c","schema_version":"1.0","event_id":"sha256:06a7f7acd633fa921b8c6b0ae85b48bb69c5d66d611bc2dcaab18e42d07a888c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/bundle.json","state_url":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-05T07:37:51Z","links":{"resolver":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ","bundle":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/bundle.json","state":"https://pith.science/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DHPJL57ND2YD4QQJBMSH24PIKJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DHPJL57ND2YD4QQJBMSH24PIKJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a69897889539c1483763e593e5ee819e50ce205e79ee4850ebccc7ac6d10be35","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-21T21:00:47Z","title_canon_sha256":"c855d2e89bd87dc7798c8a7e516f007f1ee59044d241ceb94bdac7e182c4c2cf"},"schema_version":"1.0","source":{"id":"2410.16509","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.16509","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"arxiv_version","alias_value":"2410.16509v1","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.16509","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"pith_short_12","alias_value":"DHPJL57ND2YD","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"pith_short_16","alias_value":"DHPJL57ND2YD4QQJ","created_at":"2026-07-05T09:24:02Z"},{"alias_kind":"pith_short_8","alias_value":"DHPJL57N","created_at":"2026-07-05T09:24:02Z"}],"graph_snapshots":[{"event_id":"sha256:06a7f7acd633fa921b8c6b0ae85b48bb69c5d66d611bc2dcaab18e42d07a888c","target":"graph","created_at":"2026-07-05T09:24:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2410.16509/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Firas Trabelsi, Hamid Dadkhahi, Jiaming Luo, Lily H. Zhang, Mara Finkelstein, Markus Freitag","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-21T21:00:47Z","title":"Learning from others' mistakes: Finetuning machine translation models with span-level error annotations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.16509","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b85e2516e1e33d95da420977a5734081ea31a056dd8c8c5d9ab003bc63ca03c6","target":"record","created_at":"2026-07-05T09:24:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a69897889539c1483763e593e5ee819e50ce205e79ee4850ebccc7ac6d10be35","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-21T21:00:47Z","title_canon_sha256":"c855d2e89bd87dc7798c8a7e516f007f1ee59044d241ceb94bdac7e182c4c2cf"},"schema_version":"1.0","source":{"id":"2410.16509","kind":"arxiv","version":1}},"canonical_sha256":"19de95f7ed1eb03e42090b247d71e852694dfcbcd9799b330fe6454fee347c79","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"19de95f7ed1eb03e42090b247d71e852694dfcbcd9799b330fe6454fee347c79","first_computed_at":"2026-07-05T09:24:02.713938Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:24:02.713938Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"H5oaHVjShGAVsF70kgB4oQjr0MfUXwp4ULbHAyWGfmvPwsABXnSeYO+ZxJj3tL+OvMBuU0Q4FPoWEi8tuyblAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:24:02.714396Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.16509","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b85e2516e1e33d95da420977a5734081ea31a056dd8c8c5d9ab003bc63ca03c6","sha256:06a7f7acd633fa921b8c6b0ae85b48bb69c5d66d611bc2dcaab18e42d07a888c"],"state_sha256":"4e2a24b5fd00e02bf3de90fb4b11af5e10bd28dffc74842b04d750cf6a70d06f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u/D+Dl6VZz54qVOrVLVavqg+XEWYW1vDXSAplOwI1AHBTAhs5+L8w8SEt957ZTLfFNI/JPwD6nn6Fv/GkmZYCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T07:37:51.339941Z","bundle_sha256":"4ab1c8e40ce11ee2ee8eaeb7ecfa942d95d65a17ae71960afa107a982c6c7d16"}}