{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003276131,0.00006271579,0.0001144831,0.00009537164,0.0001742707,0.00006539264,0.0002590641,0.00001384617,0.00007977611],"category_scores_gemma":[0.0000322159,0.00005213467,0.00001891411,0.0003862883,0.000007821576,0.0002229875,0.0001164336,0.0001247625,0.00005660396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006007842,"about_ca_system_score_gemma":0.00005872815,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000656362,"about_ca_topic_score_gemma":0.0003288626,"domain_scores_codex":[0.9991454,0.0001297869,0.0001266173,0.0002055663,0.0002090514,0.0001835041],"domain_scores_gemma":[0.9995645,0.0001456462,0.00004833838,0.0001774636,0.00002613574,0.00003794676],"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":[0.0001063374,0.0000730132,0.03084537,0.000005441889,0.00001252095,0.00008183545,0.002457073,0.9163837,0.0002753141,0.02324664,0.0003420259,0.02617069],"study_design_scores_gemma":[0.002512973,0.0004781083,0.002683067,0.000005272385,0.000001861386,0.00002345183,0.0008992304,0.9649615,0.0001750848,0.0007635296,0.02727449,0.0002214311],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2192229,0.0004451854,0.6694869,0.0008500774,0.0008946505,0.0007826559,0.00000302883,0.0005635828,0.107751],"genre_scores_gemma":[0.997489,2.832635e-7,0.0008208612,0.0002638322,0.00002835693,0.00003887192,0.000004995488,0.000004516374,0.001349245],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7782661,"threshold_uncertainty_score":0.212599,"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."}}