{"id":"W4409364374","doi":"10.1609/aaai.v39i16.33918","title":"Enhance Vision-Language Alignment with Noise","year":2025,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Computer science; Noise (video); Computer vision; Artificial intelligence; Image (mathematics)","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":[0.001988366,0.001796097,0.001176877,0.0006514768,0.0006253512,0.00136734,0.002218223,0.001726373,0.002442104],"category_scores_gemma":[0.01021516,0.0005683072,0.0009217432,0.0005513503,0.001259319,0.003405241,0.003060539,0.003146562,0.00109551],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001046981,"about_ca_system_score_gemma":0.00143136,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004608662,"about_ca_topic_score_gemma":0.005369558,"domain_scores_codex":[0.9989761,0.0002329037,0.00003709684,0.00041895,0.0001990851,0.0001357912],"domain_scores_gemma":[0.9984597,0.0007048565,0.0001405303,0.0003078515,0.0002641959,0.0001227488],"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":[0.000785639,0.0006007099,0.00456374,0.0003518442,0.0002375034,0.0003235789,0.0005140876,0.5083686,0.06001357,0.02557206,0.009892175,0.3887765],"study_design_scores_gemma":[0.00002045519,0.0000743781,0.0002953796,0.000014328,0.00002215717,0.00004964303,0.00002729878,0.9787472,0.01073682,0.008918301,0.001078359,0.00001570442],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04987692,0.0007105251,0.9419777,0.0005465699,0.0001885611,0.0000977899,0.0002201568,0.003583451,0.002798316],"genre_scores_gemma":[0.7412946,0.0004219839,0.2464087,0.001467088,0.0002337551,0.0002685181,0.001345953,0.0007637276,0.007795665],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004608662,"threshold_uncertainty_score":0.01051563,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01975103152251084,"score_gpt":0.3212355657977287,"score_spread":0.3014845342752179,"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."}}