{"id":"W4385574268","doi":"10.18653/v1/2022.findings-emnlp.143","title":"How sensitive are translation systems to extra contexts? Mitigating gender bias in Neural Machine Translation models through relevant contexts.","year":2022,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Machine translation; Computer science; Inference; Metric (unit); Natural language processing; Artificial intelligence; Machine learning; Transformer; Translation (biology)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006421431,0.001085326,0.0007100188,0.0007933861,0.0007166823,0.001719333,0.0009949778,0.001191422,0.001734428],"category_scores_gemma":[0.03968146,0.0004797298,0.0005092812,0.0008481308,0.0009764731,0.003288189,0.001945627,0.001812491,0.001280784],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007783934,"about_ca_system_score_gemma":0.001177209,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003781646,"about_ca_topic_score_gemma":0.007219898,"domain_scores_codex":[0.9938949,0.003897754,0.0003289166,0.0009918717,0.0006313077,0.0002552992],"domain_scores_gemma":[0.9850118,0.008957896,0.0009378073,0.003242043,0.001569096,0.0002812719],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002326266,0.0002842296,0.07533181,0.001219489,0.0009123289,0.0006849816,0.001701319,0.1504786,0.08185249,0.01161023,0.01203646,0.6615618],"study_design_scores_gemma":[0.0002182324,0.001053038,0.02291252,0.0002706307,0.0005921483,0.001030863,0.0009405041,0.7718422,0.1265835,0.05408346,0.02030285,0.0001701285],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.7521416,0.007192689,0.2155033,0.002977574,0.0005960864,0.0001358595,0.001709015,0.008735985,0.0110079],"genre_scores_gemma":[0.9513186,0.0004953597,0.04353276,0.0005580628,0.00008047713,0.00007657389,0.001670559,0.0006953558,0.001572313],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.006421431,"threshold_uncertainty_score":0.03396016,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08517361896420383,"score_gpt":0.2946986273082353,"score_spread":0.2095250083440315,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}