{"id":"W4386576803","doi":"10.18653/v1/2023.findings-eacl.106","title":"More Robust Schema-Guided Dialogue State Tracking via Tree-Based Paraphrase Ranking","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Toyota Motor Europe; Atomic Energy of Canada Limited; Department of Science and Technology, Ministry of Science and Technology, India","keywords":"Computer science; Schema (genetic algorithms); Artificial intelligence; Natural language processing; Paraphrase; Scalability; Natural language understanding; Robustness (evolution); Machine learning; Information retrieval; Natural language; Database","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003926919,0.001342551,0.001376242,0.001414048,0.0005887235,0.002664166,0.002474044,0.001559646,0.004401859],"category_scores_gemma":[0.01460103,0.0004526812,0.0009127668,0.001306804,0.0004585946,0.004233696,0.002219104,0.002418556,0.004459696],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007163983,"about_ca_system_score_gemma":0.00158349,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005179103,"about_ca_topic_score_gemma":0.007798447,"domain_scores_codex":[0.9961436,0.001982739,0.000195003,0.001045395,0.0004592168,0.0001740092],"domain_scores_gemma":[0.9938147,0.00298627,0.0002579136,0.001837851,0.0008888455,0.0002144092],"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":[0.0008205806,0.00104279,0.005743617,0.0007476389,0.0003160383,0.0001826945,0.001055643,0.1132464,0.04468829,0.007011954,0.03718787,0.7879566],"study_design_scores_gemma":[0.0001187071,0.0002530614,0.001031421,0.00003244925,0.00005482779,0.0001139274,0.0002424529,0.9644942,0.01907268,0.007139248,0.007397844,0.00004918755],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.12671,0.001466349,0.8111115,0.0007358071,0.0003413285,0.0004521117,0.003656914,0.04687874,0.008647252],"genre_scores_gemma":[0.5442019,0.000301923,0.4345913,0.0003856625,0.00009625608,0.0003370376,0.01327073,0.001463042,0.005352162],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005179103,"threshold_uncertainty_score":0.02076775,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06953959777686648,"score_gpt":0.2812464530772721,"score_spread":0.2117068553004056,"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."}}