{"id":"W7131073923","doi":"10.1109/iccvw69036.2025.00614","title":"STORM: Token-Efficient Long Video Understanding for Multimodal LLMs","year":2025,"lang":"","type":"article","venue":"","topic":"Video Analysis and Summarization","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Security token; Encoder; Inference; Latency (audio); Key (lock); Encoding (memory); Computation; Low latency (capital markets)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000722133,0.001277378,0.0009737683,0.0007852149,0.0004011088,0.0009562962,0.00211592,0.0009881303,0.007079816],"category_scores_gemma":[0.002892217,0.0004199008,0.001066508,0.0006082547,0.0004764091,0.002807652,0.001957411,0.001689444,0.002613364],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000980828,"about_ca_system_score_gemma":0.001182827,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007840109,"about_ca_topic_score_gemma":0.01352459,"domain_scores_codex":[0.9995529,0.00008370297,0.00002789478,0.0001591642,0.0001158473,0.00006054874],"domain_scores_gemma":[0.99946,0.0002229391,0.00005371865,0.0001116342,0.0001140668,0.00003766442],"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":[0.0006387297,0.0001918233,0.0009301838,0.0003683955,0.0001900897,0.0003754369,0.0003405158,0.1183881,0.04302707,0.0186205,0.02189231,0.7950369],"study_design_scores_gemma":[0.00002521611,0.00009050609,0.0002309468,0.0000221608,0.00003129039,0.00008094304,0.00008022426,0.9631982,0.01399927,0.01629044,0.005926485,0.00002417745],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008433713,0.0004854257,0.9746114,0.0001761289,0.00008037641,0.00009454096,0.0007901819,0.01396745,0.001360748],"genre_scores_gemma":[0.2862349,0.0006985586,0.6937041,0.0005498877,0.0001491482,0.0004403956,0.0062316,0.001580983,0.01041045],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007840109,"threshold_uncertainty_score":0.02368432,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03637450409026498,"score_gpt":0.2849731002703537,"score_spread":0.2485985961800887,"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."}}