{"id":"W4409366480","doi":"10.1609/aaai.v39i7.32768","title":"Leveraging Large Vision-Language Model as User Intent-Aware Encoder for Composed Image Retrieval","year":2025,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute of Aging","funders":"National Natural Science Foundation of China","keywords":"Computer science; Encoder; Computer vision; Image (mathematics); Artificial intelligence; Image retrieval; Information retrieval; Language model; Natural language processing","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":[],"consensus_categories":[],"category_scores_codex":[0.0007664812,0.0002764467,0.0003249708,0.0002289776,0.0003592239,0.0004481113,0.002407649,0.0001326438,0.00003649886],"category_scores_gemma":[0.0006431241,0.0002139238,0.0002128442,0.000864016,0.0002315725,0.0006332426,0.0005942013,0.0003395017,0.0000369141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009001967,"about_ca_system_score_gemma":0.0002302072,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001401314,"about_ca_topic_score_gemma":0.000001585726,"domain_scores_codex":[0.9977574,0.00002049294,0.0006474556,0.0006529148,0.0004803308,0.0004414163],"domain_scores_gemma":[0.9975455,0.0001463695,0.0003532152,0.0004584343,0.001416644,0.00007987513],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001439688,0.0001894298,0.00001812424,0.00008120466,0.00001754236,3.028608e-7,0.001287564,0.000008079337,0.3096987,0.6746902,0.000505571,0.01335928],"study_design_scores_gemma":[0.00004546423,0.0000764917,0.00001512828,0.0001982781,0.00001044669,0.000001065746,0.0006678808,0.3194656,0.5803204,0.09893021,0.0001252131,0.0001438172],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01572406,0.00002806374,0.9711357,0.006459924,0.0002787217,0.0008081136,0.00001501596,0.0002513438,0.005299066],"genre_scores_gemma":[0.976728,0.0000282011,0.01978693,0.0008862891,0.00002994305,0.00004334469,0.000002179411,0.0000154675,0.002479705],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9610039,"threshold_uncertainty_score":0.8723561,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04652793084147333,"score_gpt":0.3424371734148765,"score_spread":0.2959092425734031,"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."}}