{"id":"W2986143601","doi":"10.18653/v1/k19-1020","title":"Fully Unsupervised Crosslingual Semantic Textual Similarity Metric Based on BERT for Identifying Parallel Data","year":2019,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"Computer science; Semantic similarity; Artificial intelligence; Metric (unit); Similarity (geometry); Natural language processing; Information retrieval; Pattern recognition (psychology)","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001012476,0.0002143804,0.0002810365,0.0002593553,0.0001731387,0.0004739496,0.002797704,0.0001052075,0.00008831016],"category_scores_gemma":[0.0002977997,0.0001923436,0.00009723964,0.0005094633,0.00002106828,0.0007278048,0.000780753,0.0001859772,0.0001467317],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004774571,"about_ca_system_score_gemma":0.0001811613,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001418353,"about_ca_topic_score_gemma":0.0000608259,"domain_scores_codex":[0.9974682,0.00005928085,0.0003876257,0.001056676,0.0005389375,0.0004892837],"domain_scores_gemma":[0.9965508,0.0005565119,0.00008619405,0.002570429,0.0001196254,0.0001164438],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005716267,0.002512676,0.05251691,0.002110736,0.0004707345,0.000149417,0.002944028,0.3625866,0.006246655,0.2944676,0.01701123,0.2584118],"study_design_scores_gemma":[0.001228989,0.00009176575,0.001260195,0.00002852898,0.0000122275,0.00000236242,0.0000396049,0.9946271,0.0003037494,0.0007948546,0.001344611,0.000265986],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03693142,0.00004171688,0.9588498,0.0008552629,0.0005421985,0.0005949388,0.00001883188,0.0002400089,0.001925749],"genre_scores_gemma":[0.7477927,0.000001430975,0.2499081,0.001379907,0.00009059296,0.00001106808,0.00004396812,0.00001532709,0.000756874],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7108613,"threshold_uncertainty_score":0.7843546,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1050365625364869,"score_gpt":0.3355009182367941,"score_spread":0.2304643557003073,"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."}}