{"id":"W4390632530","doi":"10.48550/arxiv.2401.02297","title":"Are LLMs Robust for Spoken Dialogues?","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; McGill University","keywords":"Perplexity; Robustness (evolution); Spoken language; Computer science; Task (project management); Natural language processing; Artificial intelligence; Set (abstract data type); Speech recognition; Language model","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002460327,0.0003053937,0.000340524,0.0002591996,0.0001135859,0.0002228881,0.00207789,0.0003177172,0.00001160928],"category_scores_gemma":[0.00005019871,0.0003493932,0.0003130748,0.0003339263,0.00005076952,0.0001869651,0.003577714,0.0005756195,0.0001555306],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002191017,"about_ca_system_score_gemma":0.0001653697,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008401223,"about_ca_topic_score_gemma":0.00007957109,"domain_scores_codex":[0.9978349,0.00005120958,0.0001944843,0.001444368,0.00008405017,0.0003909599],"domain_scores_gemma":[0.997974,0.0001045764,0.00023552,0.001403346,0.0001345912,0.0001479927],"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":[0.00001373052,0.00004492177,0.0007069705,0.0003676423,0.0001104971,0.0004597519,0.0002745672,0.6061226,0.000009384271,0.3887016,0.00209636,0.00109206],"study_design_scores_gemma":[0.000201815,0.0000155111,0.0001775302,0.0001579229,0.00006197513,0.000002758773,0.00003794779,0.7934998,0.00003465127,0.2037543,0.001693412,0.0003623743],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07447043,0.000159402,0.9197397,0.0003649555,0.002262692,0.0004368155,0.00004435686,0.0005116501,0.00201006],"genre_scores_gemma":[0.9831314,0.00004773817,0.01186982,0.0001648979,0.0003330781,0.000003573078,0.00001581169,0.00002881241,0.004404874],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9086609,"threshold_uncertainty_score":0.9998958,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.17638339622166,"score_gpt":0.1969008740361831,"score_spread":0.02051747781452307,"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."}}