{"id":"W4415247602","doi":"10.48550/arxiv.2505.03176","title":"seq-JEPA: Autoregressive Predictive Learning of Invariant-Equivariant World Models","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds de Recherche du Québec - Santé; Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Institut de Valorisation des Données; Canada Excellence Research Chairs, Government of Canada; Canadian Institute for Advanced Research","keywords":"Embedding; Feature learning; Autoregressive model; Representation (politics); Multi-task learning; Inference; Encoder; Aggregate (composite); Flexibility (engineering); Transformation (genetics)","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":[0.001606976,0.0013824,0.001029532,0.0006089636,0.0003190901,0.001102893,0.003685508,0.001148242,0.002719756],"category_scores_gemma":[0.004115642,0.0009608374,0.001447091,0.0007453866,0.000945875,0.002494107,0.002284695,0.003856723,0.001183529],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007040885,"about_ca_system_score_gemma":0.001055362,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004706449,"about_ca_topic_score_gemma":0.007934484,"domain_scores_codex":[0.9993175,0.0001911075,0.00002867059,0.0002696505,0.0001248636,0.00006811517],"domain_scores_gemma":[0.9984395,0.0006911579,0.0001450566,0.0004185669,0.0002299933,0.00007569173],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001217276,0.0001837569,0.001295095,0.00009987337,0.000199475,0.0001308978,0.0001079959,0.7360524,0.004799096,0.0141821,0.006523466,0.2363041],"study_design_scores_gemma":[0.000003390153,0.00001529411,0.00005581953,0.000002524736,0.000004879772,0.000008528267,0.000002730471,0.9941518,0.000607387,0.004882718,0.0002611027,0.000003852686],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.012164,0.0002157685,0.9825397,0.0001778528,0.00005028698,0.00004373418,0.0001633667,0.003632795,0.001012499],"genre_scores_gemma":[0.5994707,0.0004619188,0.3884947,0.0007730429,0.0001366358,0.000285021,0.002563616,0.0009007143,0.006913706],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004706449,"threshold_uncertainty_score":0.009358108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06392914997216337,"score_gpt":0.2800418428867269,"score_spread":0.2161126929145635,"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."}}