{"id":"W4393147607","doi":"10.1609/aaai.v38i14.29497","title":"Lost Domain Generalization Is a Natural Consequence of Lack of Training Domains","year":2024,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Generalization; Natural (archaeology); Training (meteorology); Domain (mathematical analysis); Computer science; Psychology; Artificial intelligence; Mathematics; History; Geography","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":[0.0006816164,0.0001993573,0.0003125246,0.0001755241,0.00009517407,0.0001750361,0.001261088,0.00008325978,0.0001012088],"category_scores_gemma":[0.0002987615,0.000156659,0.0001583822,0.001137902,0.0004880738,0.000463292,0.0001970572,0.0002796867,0.00003665272],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003983783,"about_ca_system_score_gemma":0.0001982139,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003183216,"about_ca_topic_score_gemma":0.00000621628,"domain_scores_codex":[0.9979232,0.00003375427,0.0007415384,0.0004379932,0.0005924614,0.0002710836],"domain_scores_gemma":[0.9985769,0.000158674,0.0003461774,0.0002413825,0.0006120004,0.00006486881],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00002671382,0.00003485012,0.00003488453,0.0001119907,0.00002139207,0.000001578946,0.008775211,0.0001108243,0.2016097,0.762357,0.00006240192,0.02685346],"study_design_scores_gemma":[0.00004245863,0.0001606114,0.0000912545,0.0007782392,0.00001573841,0.00002155656,0.002074615,0.1927504,0.6300474,0.1734197,0.0003794334,0.0002185199],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6808285,0.0003454947,0.2910948,0.006931075,0.001260744,0.0006782064,0.00002388933,0.0001740846,0.01866331],"genre_scores_gemma":[0.9874591,0.00005029406,0.01192777,0.0002485045,0.00003918905,0.000007160914,7.636119e-7,0.00001280154,0.0002544652],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5889373,"threshold_uncertainty_score":0.6388367,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1468599123278508,"score_gpt":0.3403925985309042,"score_spread":0.1935326862030534,"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."}}