{"id":"W4404782721","doi":"10.18653/v1/2024.emnlp-main.373","title":"Unifying Multimodal Retrieval via Document Screenshot Embedding","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Information retrieval; Embedding; Document retrieval; Artificial intelligence; Natural language processing","routes":{"ca_aff":true,"ca_fund":true,"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":[0.00039203,0.0001194291,0.0001033535,0.0001216423,0.00009026277,0.0004664882,0.0005362573,0.00005142372,0.0001150807],"category_scores_gemma":[0.00002011039,0.0001027282,0.00006551703,0.0003736632,0.0000140675,0.0006280789,0.0003465871,0.0001751135,0.0001931115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008539058,"about_ca_system_score_gemma":0.00005204946,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008498762,"about_ca_topic_score_gemma":0.000004299447,"domain_scores_codex":[0.9986762,0.00003026957,0.0002120443,0.0004693514,0.0003195014,0.0002925979],"domain_scores_gemma":[0.9993288,0.00008826541,0.00001727473,0.0004423628,0.0000311297,0.00009214816],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000009839378,0.00002914145,0.0001664786,0.0001176743,0.00007531893,0.0003052524,0.002149792,0.01413028,0.01198259,0.571874,0.001067804,0.3980918],"study_design_scores_gemma":[0.00008652611,0.00001765064,0.00003542022,0.00004426936,0.000003545191,0.00002502809,0.0000241556,0.9883266,0.003297501,0.005368373,0.00262701,0.0001438904],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01166276,0.0002696854,0.9779512,0.001173386,0.0008301957,0.00009714127,2.782153e-7,0.0007056952,0.007309615],"genre_scores_gemma":[0.7354734,0.000005405749,0.2628034,0.0001920183,0.000133068,0.000001679752,6.329643e-7,0.000008105116,0.00138226],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9741964,"threshold_uncertainty_score":0.4498357,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02529742333990012,"score_gpt":0.3007426239319106,"score_spread":0.2754452005920105,"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."}}