{"id":"W7131068881","doi":"10.1109/iccvw69036.2025.00121","title":"MedVisionLlama: Leveraging Pre-Trained Large Language Model Layers to Enhance Medical Image Segmentation","year":2025,"lang":"","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Montreal Neurological Institute and Hospital","funders":"","keywords":"Segmentation; Image segmentation; Encoder; Scale-space segmentation; Segmentation-based object categorization; Feature (linguistics); Generalization; Scalability","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":[0.001693711,0.0004482561,0.0004264263,0.0004437092,0.0006127075,0.0004882115,0.002090117,0.0002385261,0.0009116362],"category_scores_gemma":[0.0008867178,0.0004492396,0.0001602853,0.001704098,0.00008948991,0.0005921172,0.001462245,0.000750491,0.0004794589],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002691081,"about_ca_system_score_gemma":0.0007646031,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008869834,"about_ca_topic_score_gemma":0.0001674585,"domain_scores_codex":[0.9952041,0.0002959335,0.0008585626,0.001475506,0.001269049,0.0008968859],"domain_scores_gemma":[0.997224,0.000393072,0.0001837983,0.001403526,0.0002455104,0.0005500562],"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":[0.000104735,0.001441625,0.001081939,0.0003957517,0.0002388116,0.00005840193,0.05402619,0.1001691,0.1147174,0.07436544,0.01233047,0.6410701],"study_design_scores_gemma":[0.0007008685,0.0000427717,0.003339508,0.0002123911,0.0000281111,0.00000361601,0.0004508588,0.9837437,0.01008017,0.0007031236,0.0002794149,0.0004154733],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1111169,0.0001137614,0.8424528,0.03545017,0.0003240558,0.0009681489,0.0000166369,0.0004340609,0.009123502],"genre_scores_gemma":[0.7441481,0.00001849071,0.2419124,0.00622679,0.00007355612,0.0001550678,0.00002019646,0.00002440411,0.007420892],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8835745,"threshold_uncertainty_score":0.9997959,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006658952244160694,"score_gpt":0.3572847019325188,"score_spread":0.3506257496883581,"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."}}