{"id":"W4405174106","doi":"10.1109/iccv51701.2025.01497","title":"LoRA.rar: Learning to Merge LoRAs via Hypernetworks for Subject-Style Conditioned Image Generation","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian University Press; European Commission","keywords":"Merge (version control); Computer science; Artificial intelligence; Style (visual arts); Art; Information retrieval; Visual arts","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005926767,0.0003342117,0.0003802043,0.0002823189,0.0002945398,0.000586572,0.00107895,0.0003426108,0.00009540442],"category_scores_gemma":[0.0001519203,0.0003312136,0.0002191536,0.000459589,0.00003904464,0.0003599377,0.0009265973,0.000482957,0.00008616866],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001791565,"about_ca_system_score_gemma":0.0002338377,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000452099,"about_ca_topic_score_gemma":0.00001469859,"domain_scores_codex":[0.9977103,0.0001342192,0.0005182911,0.0009732745,0.0003076056,0.000356242],"domain_scores_gemma":[0.9980546,0.0001395064,0.0002448436,0.0008110974,0.0006228302,0.000127106],"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":[0.0001345498,0.0005966875,0.0001478732,0.0009924886,0.0004039843,0.00001893953,0.002519072,0.005418127,0.1713567,0.1938957,0.2374772,0.3870387],"study_design_scores_gemma":[0.0002494844,0.0001271348,0.00007032754,0.00008205448,0.00003493876,0.000002821056,0.00002924703,0.7485274,0.2122141,0.006754154,0.03127022,0.0006380499],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0002441021,0.000127431,0.9885266,0.00240131,0.0008053813,0.001244794,0.0000274438,0.001090467,0.005532509],"genre_scores_gemma":[0.1980881,0.0001462142,0.7595117,0.00168461,0.0006810984,0.001354126,0.0008431773,0.00004217387,0.03764875],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7431093,"threshold_uncertainty_score":0.999914,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02716344502001248,"score_gpt":0.294603943127127,"score_spread":0.2674404981071145,"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."}}