{"id":"W4391871827","doi":"10.48550/arxiv.2402.08532","title":"Captions Are Worth a Thousand Words: Enhancing Product Retrieval with Pretrained Image-to-Text Models","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Mitacs","keywords":"Product (mathematics); Computer science; Image (mathematics); Natural language processing; Information retrieval; Artificial intelligence; Mathematics","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":[0.00109057,0.001512618,0.0007196578,0.001371275,0.0003262986,0.00119474,0.001459118,0.001548353,0.003582378],"category_scores_gemma":[0.006056451,0.0003859278,0.0009972625,0.001151311,0.0006525454,0.002698661,0.0007743111,0.001599025,0.004100256],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009405925,"about_ca_system_score_gemma":0.0005216932,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0067179,"about_ca_topic_score_gemma":0.007399934,"domain_scores_codex":[0.9995553,0.0001150473,0.00002657397,0.0001644104,0.00009484758,0.00004377148],"domain_scores_gemma":[0.9978322,0.001190021,0.0001880414,0.0003302962,0.0003915529,0.00006781826],"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":[0.00117247,0.0006828393,0.00355533,0.0007819937,0.0002231744,0.0005798609,0.0002775962,0.1577702,0.04859455,0.003204039,0.04722227,0.7359357],"study_design_scores_gemma":[0.00004690029,0.0002558881,0.001239788,0.00004394951,0.00008493761,0.000258447,0.00005069965,0.9704206,0.01835718,0.002574705,0.006631611,0.00003512364],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2459103,0.007256508,0.6939797,0.002592399,0.00113896,0.000789753,0.005989031,0.02338215,0.01896112],"genre_scores_gemma":[0.6948318,0.00215234,0.2654546,0.002191912,0.0007586343,0.0005374482,0.0139348,0.001144018,0.01899442],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0067179,"threshold_uncertainty_score":0.01335758,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03925230774777872,"score_gpt":0.2080312265532537,"score_spread":0.168778918805475,"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."}}