{"id":"W4385573049","doi":"10.18653/v1/2022.emnlp-main.195","title":"Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning","year":2022,"lang":"en","type":"article","venue":"","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Canadian Institute for Advanced Research","funders":"Hong Kong Polytechnic University; Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China; Canadian Institute for Advanced Research","keywords":"Conversation; Computer science; Heuristics; Turn-taking; Facial expression; Human–computer interaction; Term (time); Artificial intelligence; Psychology","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001106859,0.001431588,0.000932481,0.000598231,0.0004443234,0.001013463,0.001348726,0.001077294,0.003703926],"category_scores_gemma":[0.004123257,0.0004150217,0.0005880197,0.000314083,0.0003455135,0.001291223,0.00138749,0.001198493,0.001330068],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004709508,"about_ca_system_score_gemma":0.001067276,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004085712,"about_ca_topic_score_gemma":0.004590082,"domain_scores_codex":[0.999243,0.0002277525,0.00005125827,0.0002467613,0.0001570564,0.00007426273],"domain_scores_gemma":[0.9984817,0.000896384,0.0000745123,0.0001637315,0.0002621202,0.0001215698],"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":[0.001703628,0.001068437,0.003596719,0.0006325261,0.0001956066,0.0004290104,0.001281727,0.1099934,0.08675262,0.003663784,0.0114077,0.7792748],"study_design_scores_gemma":[0.0001450916,0.0002887809,0.0007201242,0.00001780693,0.00006102648,0.0001349685,0.0001569494,0.9763638,0.01587135,0.001930683,0.004266212,0.00004315429],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1176241,0.001544558,0.8506243,0.0004888398,0.0002634106,0.0005041447,0.0004055663,0.02193045,0.006614702],"genre_scores_gemma":[0.6623493,0.0002461506,0.3321808,0.0003049431,0.00006308655,0.0003460636,0.0008629353,0.0004085535,0.003238155],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004085712,"threshold_uncertainty_score":0.01239085,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04671185926340579,"score_gpt":0.2549916835519332,"score_spread":0.2082798242885274,"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."}}