{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:LXVRCL5EXYPAGGLHF2SOHBE2CE","short_pith_number":"pith:LXVRCL5E","schema_version":"1.0","canonical_sha256":"5deb112fa4be1e0319672ea4e3849a1111552a66342e1f5cb4cb2cd863d4642e","source":{"kind":"arxiv","id":"1508.02445","version":1},"attestation_state":"computed","paper":{"title":"Removing Biases from Trainable MT Metrics by Using Self-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Milo\\v{s} Stanojevi\\'c","submitted_at":"2015-08-10T22:24:36Z","abstract_excerpt":"Most trainable machine translation (MT) metrics train their weights on human judgments of state-of-the-art MT systems outputs. This makes trainable metrics biases in many ways. One of them is preferring longer translations. These biased metrics when used for tuning are evaluating different types of translations -- n-best lists of translations with very diverse quality. Systems tuned with these metrics tend to produce overly long translations that are preferred by the metric but not by humans. This is usually solved by manually tweaking metric's weights to equally value recall and precision. Ou"},"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":"1508.02445","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2015-08-10T22:24:36Z","cross_cats_sorted":[],"title_canon_sha256":"fd6312c758579f1478ee0cec22c6601e1a7872f65f80d86d7db22b9de5c4e6e6","abstract_canon_sha256":"b7f856a5d5f8ad86a03a04c8227408a0a15338a2aa474febeed12d9d7a9e5332"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:35:29.209808Z","signature_b64":"IbGvrR0RvxRjboeW1tu0CR/pQT10GVMBrKhF5VfZOX3oXy/ey8PgUogxQQP4YrBzdOeTzZ8Q6sZQ++c8LLNlAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5deb112fa4be1e0319672ea4e3849a1111552a66342e1f5cb4cb2cd863d4642e","last_reissued_at":"2026-05-18T01:35:29.208937Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:35:29.208937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Removing Biases from Trainable MT Metrics by Using Self-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Milo\\v{s} Stanojevi\\'c","submitted_at":"2015-08-10T22:24:36Z","abstract_excerpt":"Most trainable machine translation (MT) metrics train their weights on human judgments of state-of-the-art MT systems outputs. This makes trainable metrics biases in many ways. One of them is preferring longer translations. These biased metrics when used for tuning are evaluating different types of translations -- n-best lists of translations with very diverse quality. Systems tuned with these metrics tend to produce overly long translations that are preferred by the metric but not by humans. This is usually solved by manually tweaking metric's weights to equally value recall and precision. Ou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1508.02445","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":""},"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":"1508.02445","created_at":"2026-05-18T01:35:29.209096+00:00"},{"alias_kind":"arxiv_version","alias_value":"1508.02445v1","created_at":"2026-05-18T01:35:29.209096+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1508.02445","created_at":"2026-05-18T01:35:29.209096+00:00"},{"alias_kind":"pith_short_12","alias_value":"LXVRCL5EXYPA","created_at":"2026-05-18T12:29:29.992203+00:00"},{"alias_kind":"pith_short_16","alias_value":"LXVRCL5EXYPAGGLH","created_at":"2026-05-18T12:29:29.992203+00:00"},{"alias_kind":"pith_short_8","alias_value":"LXVRCL5E","created_at":"2026-05-18T12:29:29.992203+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/LXVRCL5EXYPAGGLHF2SOHBE2CE","json":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE.json","graph_json":"https://pith.science/api/pith-number/LXVRCL5EXYPAGGLHF2SOHBE2CE/graph.json","events_json":"https://pith.science/api/pith-number/LXVRCL5EXYPAGGLHF2SOHBE2CE/events.json","paper":"https://pith.science/paper/LXVRCL5E"},"agent_actions":{"view_html":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE","download_json":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE.json","view_paper":"https://pith.science/paper/LXVRCL5E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1508.02445&json=true","fetch_graph":"https://pith.science/api/pith-number/LXVRCL5EXYPAGGLHF2SOHBE2CE/graph.json","fetch_events":"https://pith.science/api/pith-number/LXVRCL5EXYPAGGLHF2SOHBE2CE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE/action/storage_attestation","attest_author":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE/action/author_attestation","sign_citation":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE/action/citation_signature","submit_replication":"https://pith.science/pith/LXVRCL5EXYPAGGLHF2SOHBE2CE/action/replication_record"}},"created_at":"2026-05-18T01:35:29.209096+00:00","updated_at":"2026-05-18T01:35:29.209096+00:00"}