{"id":"W4385570249","doi":"10.18653/v1/2023.findings-acl.177","title":"What Knowledge Is Needed? Towards Explainable Memory for kNN-MT Domain Adaptation","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Novelis (Canada)","funders":"National Natural Science Foundation of China; National Science Foundation","keywords":"Correctness; Computer science; Interpretability; Security token; Artificial intelligence; Adaptation (eye); Pruning; Domain (mathematical analysis); Domain adaptation; Machine translation; Natural language processing; Machine learning; Theoretical computer science; Algorithm","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":[0.004070393,0.0009433455,0.001292126,0.001270219,0.0008562453,0.002859185,0.002552968,0.001418261,0.003012188],"category_scores_gemma":[0.03460214,0.0007938981,0.000929613,0.001221047,0.001915253,0.009432675,0.003845268,0.002843852,0.00104584],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001357895,"about_ca_system_score_gemma":0.001919415,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004223009,"about_ca_topic_score_gemma":0.005335565,"domain_scores_codex":[0.9974176,0.001201594,0.0002191858,0.0006271459,0.0003604933,0.000173988],"domain_scores_gemma":[0.9806467,0.01078488,0.0009862133,0.005950734,0.001303366,0.000328083],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00100425,0.0002965688,0.01474039,0.0008614202,0.0003384061,0.00124339,0.006823125,0.1807466,0.02776489,0.1020578,0.00674943,0.6573737],"study_design_scores_gemma":[0.0000435805,0.00008651326,0.001387182,0.00009692686,0.0001151218,0.000266505,0.0008988654,0.8042328,0.0134438,0.1724221,0.006965744,0.00004073438],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06638227,0.0004223315,0.9265909,0.001746923,0.0000388311,0.0001010267,0.0003631557,0.002539437,0.001815101],"genre_scores_gemma":[0.696974,0.0003782464,0.2990202,0.0003569004,0.00008394547,0.000205247,0.001240166,0.0005063593,0.001234927],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004223009,"threshold_uncertainty_score":0.02152652,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06855755797737949,"score_gpt":0.3003839391310166,"score_spread":0.2318263811536371,"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."}}