{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QW22EOIHOL4DV2ILK2ABQDT7KX","short_pith_number":"pith:QW22EOIH","schema_version":"1.0","canonical_sha256":"85b5a2390772f83ae90b5680180e7f55f8b2c239e8a6eb43a18e772187991485","source":{"kind":"arxiv","id":"2412.17592","version":2},"attestation_state":"computed","paper":{"title":"Investigating Length Issues in Document-level Machine Translation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fran\\c{c}ois Yvon, Rachel Bawden, Ziqian Peng","submitted_at":"2024-12-23T14:08:45Z","abstract_excerpt":"Transformer architectures are increasingly effective at processing and generating very long chunks of texts, opening new perspectives for document-level machine translation (MT). In this work, we challenge the ability of MT systems to handle texts comprising up to several thousands of tokens. We design and implement a new approach designed to precisely measure the effect of length increments on MT outputs. Our experiments with two representative architectures unambiguously show that (a)~translation performance decreases with the length of the input text; (b)~the position of sentences within th"},"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":"2412.17592","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-23T14:08:45Z","cross_cats_sorted":[],"title_canon_sha256":"2d853709f196c753d5c02f86d0f2b69552330297e8a84723d8ff75db7c608149","abstract_canon_sha256":"0dce9634a6eed97b6f15b79ca2fc0a8462b508db4592166242838786bb3b0171"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:40.857420Z","signature_b64":"6ruh44J5oq09R8ijEqnZv/EeiLVpOPj11lLminr3PXPX9ojEqnYTVnPKp1moIPNN3qTpvb4PsZwyhw9hWG9XBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85b5a2390772f83ae90b5680180e7f55f8b2c239e8a6eb43a18e772187991485","last_reissued_at":"2026-07-05T10:54:40.856944Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:40.856944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Investigating Length Issues in Document-level Machine Translation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fran\\c{c}ois Yvon, Rachel Bawden, Ziqian Peng","submitted_at":"2024-12-23T14:08:45Z","abstract_excerpt":"Transformer architectures are increasingly effective at processing and generating very long chunks of texts, opening new perspectives for document-level machine translation (MT). In this work, we challenge the ability of MT systems to handle texts comprising up to several thousands of tokens. We design and implement a new approach designed to precisely measure the effect of length increments on MT outputs. Our experiments with two representative architectures unambiguously show that (a)~translation performance decreases with the length of the input text; (b)~the position of sentences within th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17592","kind":"arxiv","version":2},"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/2412.17592/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":"2412.17592","created_at":"2026-07-05T10:54:40.856992+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.17592v2","created_at":"2026-07-05T10:54:40.856992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17592","created_at":"2026-07-05T10:54:40.856992+00:00"},{"alias_kind":"pith_short_12","alias_value":"QW22EOIHOL4D","created_at":"2026-07-05T10:54:40.856992+00:00"},{"alias_kind":"pith_short_16","alias_value":"QW22EOIHOL4DV2IL","created_at":"2026-07-05T10:54:40.856992+00:00"},{"alias_kind":"pith_short_8","alias_value":"QW22EOIH","created_at":"2026-07-05T10:54:40.856992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02461","citing_title":"XToM: Exploring the Multilingual Theory of Mind for Large Language Models","ref_index":2022,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX","json":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX.json","graph_json":"https://pith.science/api/pith-number/QW22EOIHOL4DV2ILK2ABQDT7KX/graph.json","events_json":"https://pith.science/api/pith-number/QW22EOIHOL4DV2ILK2ABQDT7KX/events.json","paper":"https://pith.science/paper/QW22EOIH"},"agent_actions":{"view_html":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX","download_json":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX.json","view_paper":"https://pith.science/paper/QW22EOIH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.17592&json=true","fetch_graph":"https://pith.science/api/pith-number/QW22EOIHOL4DV2ILK2ABQDT7KX/graph.json","fetch_events":"https://pith.science/api/pith-number/QW22EOIHOL4DV2ILK2ABQDT7KX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX/action/storage_attestation","attest_author":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX/action/author_attestation","sign_citation":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX/action/citation_signature","submit_replication":"https://pith.science/pith/QW22EOIHOL4DV2ILK2ABQDT7KX/action/replication_record"}},"created_at":"2026-07-05T10:54:40.856992+00:00","updated_at":"2026-07-05T10:54:40.856992+00:00"}