{"id":"W4417405775","doi":"10.64898/2025.12.12.694037","title":"Towards Video-LLM Driven Workflow for Behavioral Segmentation and Scoring in Mice Performing a Skilled Water Reaching Task: An Evaluation of Recent LLM Models","year":2025,"lang":"","type":"article","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Zebrafish Biomedical Research Applications","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research","keywords":"Benchmark (surveying); Workflow; Segmentation; Key (lock); Transformative learning; Pixel; Fidelity; Interpolation (computer graphics)","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":[0.003488419,0.001065299,0.0006937535,0.0009756228,0.0002938157,0.001130206,0.00215568,0.001309252,0.002654452],"category_scores_gemma":[0.004232239,0.0004628559,0.0008678937,0.0003429021,0.0003811178,0.0008134131,0.00120991,0.001336612,0.001687045],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001112418,"about_ca_system_score_gemma":0.0009283039,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002045549,"about_ca_topic_score_gemma":0.002471587,"domain_scores_codex":[0.9991756,0.0001379304,0.00005750535,0.000175688,0.0003855265,0.00006763802],"domain_scores_gemma":[0.9981112,0.0005237054,0.000242873,0.0003462667,0.0005560554,0.0002198905],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001009736,0.0004832983,0.005067138,0.001289207,0.0002238521,0.0002877406,0.0003298958,0.02094037,0.793803,0.00247965,0.009445216,0.1646409],"study_design_scores_gemma":[0.0001096787,0.001865724,0.01082123,0.000239619,0.0002317105,0.0006537957,0.0001343619,0.4243459,0.5297365,0.001978595,0.0296487,0.0002341146],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09472969,0.001399119,0.8673945,0.0006054938,0.0002783848,0.0006051575,0.002469009,0.03056208,0.00195653],"genre_scores_gemma":[0.213524,0.001874039,0.7711955,0.0003971264,0.00006455714,0.00187144,0.004293946,0.00333228,0.00344722],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003488419,"threshold_uncertainty_score":0.01844877,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03205041988482443,"score_gpt":0.318275235782619,"score_spread":0.2862248158977946,"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."}}