{"id":"W4385574182","doi":"10.18653/v1/2022.findings-emnlp.318","title":"Controllable Dialogue Simulation with In-context Learning","year":2022,"lang":"en","type":"article","venue":"","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Waterloo","funders":"Defense Advanced Research Projects Agency","keywords":"Dialogic; Computer science; Crowdsourcing; Context (archaeology); Annotation; Fluency; Set (abstract data type); Workflow; Artificial intelligence; Training set; Language model; Code (set theory); Natural language processing; Machine learning; World Wide Web; Database; Programming language; Linguistics","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":[0.00301372,0.00218367,0.001187865,0.0007532222,0.0008598685,0.001593974,0.003280364,0.001457603,0.004845552],"category_scores_gemma":[0.01333984,0.0009263318,0.001442297,0.0004109402,0.001258377,0.003007581,0.004402675,0.002852213,0.002539598],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009384576,"about_ca_system_score_gemma":0.00140474,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002943621,"about_ca_topic_score_gemma":0.005776855,"domain_scores_codex":[0.995868,0.002333089,0.0001455233,0.001194449,0.0002939348,0.0001650678],"domain_scores_gemma":[0.9946185,0.003525176,0.000193636,0.001013741,0.0003545369,0.000294364],"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":[0.001466592,0.001254862,0.007726481,0.001311417,0.0003457379,0.0006393966,0.002706057,0.500693,0.05066367,0.01436672,0.02202906,0.3967971],"study_design_scores_gemma":[0.00008863014,0.0001485846,0.0004328763,0.00003579042,0.00002588396,0.00009069708,0.000183001,0.9711958,0.00997306,0.01049277,0.007292725,0.00004030624],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04389942,0.0006026099,0.927799,0.0003826716,0.0001966727,0.0004659116,0.000954246,0.02156666,0.004132788],"genre_scores_gemma":[0.5373822,0.0002046715,0.4519495,0.0005165374,0.00008889029,0.001517587,0.003367996,0.001129972,0.00384268],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004845552,"threshold_uncertainty_score":0.01621002,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01078697254154309,"score_gpt":0.2146872549640169,"score_spread":0.2039002824224738,"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."}}