{"id":"W4402772286","doi":"10.1109/cvpr52733.2024.01355","title":"EgoThink: Evaluating First-Person Perspective Thinking Capability of Vision-Language Models","year":2024,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Perspective (graphical); Computer science; Human–computer interaction; Artificial intelligence; Natural language processing","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":[0.0009030587,0.0001488321,0.0001733115,0.0001424975,0.0001992345,0.0002084391,0.0008214459,0.00006421169,0.00009887695],"category_scores_gemma":[0.0002554985,0.0001248014,0.00009451635,0.0005885012,0.00006154116,0.0006203791,0.0002366014,0.000321303,0.00005883013],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001923584,"about_ca_system_score_gemma":0.000113206,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006346277,"about_ca_topic_score_gemma":0.00008448296,"domain_scores_codex":[0.9982645,0.0001168999,0.0002389279,0.0006467531,0.0005179548,0.0002149929],"domain_scores_gemma":[0.9983225,0.0005926225,0.00006769788,0.0007664601,0.0001843122,0.00006633125],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000003016178,0.0000801487,0.0001299843,0.00009082896,0.00003541687,0.000004144227,0.0902968,0.05453115,0.003723088,0.8265341,0.0001415107,0.02442981],"study_design_scores_gemma":[0.00008465844,0.00007707896,0.001836755,0.00006810721,0.000009285589,0.00000572551,0.001518484,0.9384106,0.0003582834,0.05749167,0.00001176103,0.0001276008],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1537002,0.000643686,0.8161112,0.005554781,0.0001475617,0.0003544838,0.000003846005,0.0008319374,0.02265227],"genre_scores_gemma":[0.8036267,0.000001841914,0.1960558,0.00007843756,0.00003757555,0.00002702858,0.000001335455,0.0000121721,0.0001591176],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8838794,"threshold_uncertainty_score":0.9593715,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02926812550214529,"score_gpt":0.357759155973917,"score_spread":0.3284910304717717,"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."}}