{"id":"W4400375663","doi":"10.1145/3654777.3676345","title":"Improving Steering and Verification in AI-Assisted Data Analysis with Interactive Task Decomposition","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Decomposition; Computer science; Task (project management); Human–computer interaction; Systems engineering; Chemistry; 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":[0.04226379,0.001849302,0.0009795276,0.001686671,0.00113261,0.004259068,0.0043473,0.001605312,0.007947289],"category_scores_gemma":[0.2126076,0.00148013,0.0009674712,0.0009363256,0.002910023,0.006339501,0.00790648,0.002670811,0.002751144],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001505952,"about_ca_system_score_gemma":0.004632708,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002373392,"about_ca_topic_score_gemma":0.002728591,"domain_scores_codex":[0.9588317,0.02867205,0.002546293,0.005398342,0.003316047,0.001235608],"domain_scores_gemma":[0.7267587,0.2092261,0.01019904,0.03582843,0.01478397,0.003203773],"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":[0.003567825,0.00232508,0.0181693,0.002285515,0.0002086325,0.0005529094,0.04343114,0.02091172,0.09891166,0.01477358,0.007021797,0.7878408],"study_design_scores_gemma":[0.002055135,0.003841244,0.02541769,0.0009895409,0.000386018,0.001343463,0.01090584,0.6248748,0.1723441,0.09718813,0.05989642,0.0007576275],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.085936,0.0001264329,0.8961338,0.0004757017,0.00006599311,0.001446882,0.0001741239,0.01244974,0.003191306],"genre_scores_gemma":[0.319899,0.00007367605,0.6747382,0.0002915397,0.00002882918,0.001886216,0.0003211224,0.0009319567,0.001829564],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04226379,"threshold_uncertainty_score":0.223515,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02211765731294867,"score_gpt":0.308756620832746,"score_spread":0.2866389635197973,"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."}}