{"id":"W4399426318","doi":"10.1162/tacl_a_00669","title":"Source-Free Domain Adaptation for Question Answering with Masked Self-training","year":2024,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Computer science; Domain adaptation; Adaptation (eye); Domain (mathematical analysis); Question answering; Training (meteorology); Training set; Natural language processing; Artificial intelligence; Information retrieval; Speech 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002699584,0.00105239,0.0008186845,0.0007698624,0.0004545663,0.0007461662,0.001947975,0.001391542,0.00185134],"category_scores_gemma":[0.006169381,0.0005001897,0.0009997393,0.0006632974,0.0008909043,0.002378234,0.002549531,0.002318628,0.001271883],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006379754,"about_ca_system_score_gemma":0.0007149871,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002303574,"about_ca_topic_score_gemma":0.002567192,"domain_scores_codex":[0.9987387,0.0005694845,0.00005777819,0.0004301883,0.0001288874,0.00007501158],"domain_scores_gemma":[0.9966581,0.001716739,0.0001578899,0.000950499,0.000397498,0.0001191824],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007723626,0.0007603513,0.006416548,0.0003856136,0.0003211002,0.000262804,0.0009438813,0.2688917,0.05941187,0.007608148,0.01329819,0.6409274],"study_design_scores_gemma":[0.00001800668,0.00007086327,0.0006737737,0.00001336011,0.00002234712,0.00006229043,0.00004506152,0.9837277,0.008378778,0.005113182,0.001859654,0.00001496666],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06854468,0.0009427209,0.9213489,0.0003480919,0.0001267196,0.0001478116,0.0002962575,0.006584733,0.001660161],"genre_scores_gemma":[0.7512888,0.0002728077,0.2420896,0.0007692216,0.0001195058,0.000326255,0.001659821,0.0004271389,0.003046776],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002699584,"threshold_uncertainty_score":0.01427692,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01946249239711238,"score_gpt":0.2532761076822381,"score_spread":0.2338136152851257,"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."}}