{"id":"W4403666342","doi":"10.48550/arxiv.2409.08212","title":"Adaptive Language-Guided Abstraction from Contrastive Explanations","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; National Defense Science and Engineering Graduate; Open Philanthropy Project; National Science Foundation","keywords":"Abstraction; Computer science; Programming language; Linguistics; Natural language processing; Contrastive analysis; Artificial intelligence; Philosophy; Epistemology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001202586,0.0002416745,0.0002316784,0.0002347441,0.00009310678,0.0001527935,0.0009263956,0.0002410271,0.00005473971],"category_scores_gemma":[0.00003002549,0.0002843058,0.0001542608,0.0002774654,0.00004185053,0.0002789421,0.001250423,0.0007394876,0.0003085331],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003123422,"about_ca_system_score_gemma":0.0002125104,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001931697,"about_ca_topic_score_gemma":0.000227993,"domain_scores_codex":[0.9983387,0.0000775029,0.0001876462,0.001067048,0.0001023417,0.0002267372],"domain_scores_gemma":[0.9986567,0.0001614134,0.0001579953,0.0007857387,0.0001264913,0.0001116166],"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":[0.00002354584,0.00008517826,0.0001863768,0.00004299032,0.0004258977,0.001361332,0.003730329,0.5212703,0.00027541,0.4669455,0.001419698,0.004233419],"study_design_scores_gemma":[0.0002091249,0.00001275223,0.0004946674,0.0001038331,0.00008090094,0.000002958915,0.0005478708,0.8698551,0.0002282035,0.1280896,0.00008431504,0.0002907],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2026223,0.0001449267,0.7885832,0.0001691898,0.001198348,0.0002081644,0.00008546747,0.0004105715,0.00657784],"genre_scores_gemma":[0.9929492,0.00003354362,0.005348749,0.00006646738,0.0001941852,0.000001845442,0.00003928351,0.00001486612,0.001351785],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.790327,"threshold_uncertainty_score":0.9999609,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1048807175927821,"score_gpt":0.2123155390683983,"score_spread":0.1074348214756163,"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."}}