{"id":"W4385571841","doi":"10.18653/v1/2023.acl-long.758","title":"Learning New Skills after Deployment: Improving open-domain internet-driven dialogue with human feedback","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Software deployment; Computer science; The Internet; Domain (mathematical analysis); Artificial intelligence; Machine learning; World Wide Web; Software 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":[0.00549527,0.001728093,0.001381145,0.0008764463,0.0005916317,0.001400179,0.002144337,0.001849742,0.001763215],"category_scores_gemma":[0.02216627,0.0007302026,0.0007957565,0.0005817247,0.0008847104,0.003765807,0.001959868,0.003099657,0.0016887],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001064126,"about_ca_system_score_gemma":0.001317361,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008711228,"about_ca_topic_score_gemma":0.01043267,"domain_scores_codex":[0.9977887,0.001287187,0.00006794483,0.0005201508,0.000160153,0.0001759785],"domain_scores_gemma":[0.9849631,0.01143572,0.000469824,0.001567714,0.0009847746,0.0005789141],"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":[0.001492793,0.001699196,0.01423153,0.0004774525,0.000315549,0.0001834676,0.001552405,0.5547531,0.009248724,0.002894416,0.0127095,0.4004418],"study_design_scores_gemma":[0.00004733854,0.0001417369,0.0005290197,0.00001157628,0.00002092669,0.00002043079,0.00006563373,0.9953706,0.001461046,0.001786068,0.0005326408,0.00001298826],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3170451,0.002079088,0.6578631,0.001714134,0.0002335329,0.0003541766,0.0007609481,0.01473633,0.005213506],"genre_scores_gemma":[0.8947198,0.0002090603,0.09933842,0.0005662303,0.0001100937,0.0001949155,0.001526525,0.0004414352,0.002893511],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008711228,"threshold_uncertainty_score":0.02906209,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01701245785068708,"score_gpt":0.2501860809956955,"score_spread":0.2331736231450084,"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."}}