{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001185739,0.0003673131,0.0005119191,0.0006073262,0.000984308,0.001104619,0.0009523064,0.0002119449,0.000166926],"category_scores_gemma":[0.0002112243,0.0003455053,0.0004691254,0.001811717,0.0001061977,0.0004823295,0.0004805223,0.0001853897,0.00003342472],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008063646,"about_ca_system_score_gemma":0.0004131849,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002045003,"about_ca_topic_score_gemma":0.0005440671,"domain_scores_codex":[0.9966481,0.0001269542,0.0009005243,0.001072043,0.0005202048,0.0007322103],"domain_scores_gemma":[0.998042,0.0003798096,0.0002244702,0.000851922,0.000320906,0.0001808868],"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":[0.00006872993,0.0005036567,0.004795243,0.0002031453,0.0004085838,0.000007489894,0.001535388,0.1320945,0.0003500402,0.789709,0.006110433,0.06421377],"study_design_scores_gemma":[0.00112299,0.00009786496,0.001376492,0.0001407798,0.00016107,8.796225e-7,0.0005900614,0.9906555,0.0008228252,0.003209328,0.001442767,0.0003794153],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00296346,0.0005105769,0.983106,0.003598018,0.002206904,0.0007965886,0.00000807483,0.0001218982,0.006688493],"genre_scores_gemma":[0.9778633,0.00005374125,0.01355993,0.000939377,0.0001156515,0.00003448186,0.00001510732,0.00001543037,0.007403019],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9748998,"threshold_uncertainty_score":0.9999323,"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."}}