{"id":"W4402667043","doi":"10.18653/v1/2024.acl-long.37","title":"OPEx: A Component-Wise Analysis of LLM-Centric Agents in Embodied Instruction Following","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Component (thermodynamics); Embodied cognition; Computer science; Artificial intelligence; Physics","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":[0.0003704341,0.0005997284,0.0003979355,0.0007453667,0.0004928645,0.001217489,0.001024297,0.0005896605,0.0113376],"category_scores_gemma":[0.001764885,0.0002740154,0.0005521128,0.0004173226,0.0004815046,0.001526699,0.001362578,0.0007527934,0.00115573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000531121,"about_ca_system_score_gemma":0.001117195,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005736961,"about_ca_topic_score_gemma":0.004619121,"domain_scores_codex":[0.9998031,0.00003632748,0.000009793001,0.00005523037,0.00005349437,0.0000420197],"domain_scores_gemma":[0.9995841,0.0001521252,0.00004608281,0.00006796421,0.0001161249,0.00003342674],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008652488,0.0002319231,0.007962347,0.0003830028,0.00013169,0.0006144562,0.001034577,0.2328573,0.04026877,0.2744614,0.00556236,0.4356269],"study_design_scores_gemma":[0.00001084735,0.00008654277,0.002102314,0.00002292696,0.00004036875,0.00006112538,0.0001854601,0.9113113,0.009356497,0.07325483,0.003551634,0.00001603735],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05069558,0.00009572496,0.9403335,0.0001077975,0.0000333591,0.00009412682,0.0003267728,0.001857393,0.006455723],"genre_scores_gemma":[0.7177081,0.0001327471,0.2658494,0.00006681983,0.00002620917,0.0001736718,0.0006708554,0.000564897,0.01480731],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0113376,"threshold_uncertainty_score":0.0379281,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02009801514821536,"score_gpt":0.3025195860109078,"score_spread":0.2824215708626924,"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."}}