{"id":"W4412889036","doi":"10.18653/v1/2025.conll-1.35","title":"GCG-Based Artificial Languages for Evaluating Inductive Biases of Neural Language Models","year":2025,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Computer science; Artificial intelligence; Natural language processing; Artificial neural network; Inductive bias; Machine learning; Engineering; Multi-task learning; Systems engineering","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":[0.009784047,0.001268682,0.0006521742,0.002485528,0.0004171527,0.001801705,0.001623203,0.00210481,0.002768254],"category_scores_gemma":[0.05959307,0.00037756,0.000814407,0.00198666,0.001884741,0.002893341,0.002230701,0.002659558,0.0004043078],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001942334,"about_ca_system_score_gemma":0.0006778874,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002884007,"about_ca_topic_score_gemma":0.002878692,"domain_scores_codex":[0.9957618,0.002571563,0.0002998023,0.0005451356,0.0006436012,0.000178043],"domain_scores_gemma":[0.9451011,0.04684256,0.001741932,0.003897679,0.001901413,0.0005153219],"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":[0.0008894269,0.0003469807,0.03045035,0.0006820083,0.0005575259,0.0002016751,0.0006386059,0.8471336,0.006595702,0.02690383,0.002371322,0.08322897],"study_design_scores_gemma":[0.00004000707,0.0001908245,0.001797789,0.00003483214,0.00002952965,0.00006184042,0.0001060845,0.9758602,0.002658738,0.01862623,0.0005698499,0.00002403126],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.7933062,0.0007829897,0.1937762,0.0005317912,0.0001601356,0.0003808697,0.002378001,0.001731412,0.006952435],"genre_scores_gemma":[0.9349274,0.0001461013,0.06205708,0.0001758221,0.0000273324,0.0004921909,0.001614664,0.0001778072,0.000381592],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.009784047,"threshold_uncertainty_score":0.05174357,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09848115993115467,"score_gpt":0.3950496185819756,"score_spread":0.2965684586508209,"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."}}