{"id":"W7131065779","doi":"10.1109/iccvw69036.2025.00518","title":"A Signer-Invariant Conformer and Multi-Scale Fusion Transformer for Continuous Sign Language Recognition","year":2025,"lang":"","type":"article","venue":"","topic":"Hand Gesture Recognition Systems","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Transformer; Encoder; Sign language; Sentence; Language model; Generalization; Fusion; Word error rate","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":[0.001227889,0.0005075283,0.0007571819,0.0004030923,0.0004631181,0.0006228671,0.0003916055,0.0004587408,0.0002795185],"category_scores_gemma":[0.0001129418,0.0004399752,0.0002714887,0.0006242919,0.000153011,0.0007640285,0.00008245396,0.0002656788,0.0001511666],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006349626,"about_ca_system_score_gemma":0.0002800741,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002573975,"about_ca_topic_score_gemma":0.0003672944,"domain_scores_codex":[0.9966295,0.0002325287,0.001030991,0.001031823,0.0003297218,0.0007454878],"domain_scores_gemma":[0.9979174,0.0005752278,0.0002312511,0.0004412054,0.0005579689,0.0002770144],"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":[0.0001669919,0.0003221393,0.00007655374,0.0005941379,0.0001422284,0.00001268852,0.006445643,0.0000010295,0.02944461,0.001214563,0.001372976,0.9602064],"study_design_scores_gemma":[0.05048471,0.004033602,0.003154099,0.006285079,0.001475225,0.0004701128,0.02675324,0.4713069,0.3340385,0.005512199,0.09108596,0.005400374],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009963988,0.001274185,0.960754,0.001464003,0.001432646,0.003405323,0.0001923899,0.0001901854,0.02132329],"genre_scores_gemma":[0.9058468,0.0003962074,0.06643912,0.002772213,0.0002791759,0.0005686797,0.0001048564,0.00004346737,0.02354942],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9548061,"threshold_uncertainty_score":0.9998052,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02227737966417603,"score_gpt":0.273309913226476,"score_spread":0.2510325335622999,"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."}}