{"id":"W4407352310","doi":"10.2139/ssrn.5131396","title":"V-Sparse: Temporal-Spatial Visual Compression and Coarse-to-Fine Alignment for Text-Video Retrieval","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Video Analysis and Summarization","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Artificial intelligence; Video retrieval; Pattern recognition (psychology); Computer vision","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.003260068,0.0004399368,0.0006982873,0.0005008113,0.0004799436,0.0006125739,0.001195804,0.0003239339,0.000009951083],"category_scores_gemma":[0.0001926098,0.000392289,0.0003123261,0.0003893095,0.00003823108,0.0002426379,0.00138303,0.001958322,0.00000507765],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001094093,"about_ca_system_score_gemma":0.003433614,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002120341,"about_ca_topic_score_gemma":0.0009184289,"domain_scores_codex":[0.9952953,0.0002667891,0.0008573587,0.0008914569,0.0006999601,0.001989068],"domain_scores_gemma":[0.9980898,0.0001484429,0.0005796603,0.0005660745,0.0003437653,0.0002722878],"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":[0.001115567,0.0008447869,0.004373538,0.0003355095,0.0021753,0.00002216113,0.001234969,0.01695204,0.00272952,0.1828644,0.007861354,0.7794909],"study_design_scores_gemma":[0.003211548,0.001782454,0.0007478253,0.0008331233,0.00047372,0.0001199568,0.0002013887,0.6971003,0.001558225,0.2696952,0.02289304,0.001383192],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009162717,0.002118558,0.9831135,0.003934992,0.0009050186,0.0005254564,0.00002162806,0.00007306973,0.0001450144],"genre_scores_gemma":[0.9808905,0.002478494,0.01052345,0.000543563,0.0008503785,0.00004775808,0.0001255481,0.0000386304,0.004501664],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9725901,"threshold_uncertainty_score":0.9998529,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0121976333115473,"score_gpt":0.282626849278032,"score_spread":0.2704292159664847,"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."}}