{"id":"W4405622294","doi":"10.48550/arxiv.2408.12109","title":"RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México; Project 211; Natural Science Foundation of Liaoning Province; National Natural Science Foundation of China; Fundamental Research Funds for the Central Universities; Strong; Institute for Catastrophic Loss Reduction; National Science Foundation","keywords":"Preference; Computer science; Psychology; Artificial intelligence; Cognitive psychology; Economics; Microeconomics","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":[0.001589951,0.001291355,0.001331335,0.0006779956,0.0004057107,0.001069635,0.002801682,0.001693575,0.003361952],"category_scores_gemma":[0.005590124,0.0006891152,0.001099182,0.0007322395,0.0006695852,0.001983258,0.001477069,0.002705903,0.001661559],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001331565,"about_ca_system_score_gemma":0.001716683,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008393516,"about_ca_topic_score_gemma":0.01172858,"domain_scores_codex":[0.9991128,0.0003121612,0.00003977543,0.0002704218,0.0001539854,0.0001108205],"domain_scores_gemma":[0.9987721,0.0005711886,0.0001273004,0.000185995,0.0002428939,0.0001005782],"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":[0.0004495026,0.0002580612,0.001111924,0.0001994791,0.0001013388,0.0001526,0.0001113388,0.6860406,0.01066737,0.01026585,0.009976025,0.2806658],"study_design_scores_gemma":[0.00001796171,0.00004652206,0.00007339861,0.000005896672,0.000004756608,0.0000174569,0.000005692361,0.995845,0.001073076,0.002468585,0.0004332077,0.000008480573],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02635808,0.0005300134,0.9647676,0.0004817364,0.0001043794,0.0001327784,0.000409475,0.004974077,0.00224189],"genre_scores_gemma":[0.5643041,0.0003057311,0.4223988,0.0008954726,0.0001002184,0.0004403062,0.001658814,0.0008904283,0.009006165],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008393516,"threshold_uncertainty_score":0.01668936,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2234317025266224,"score_gpt":0.2407001204958311,"score_spread":0.01726841796920861,"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."}}