{"id":"W4404782669","doi":"10.18653/v1/2024.emnlp-main.944","title":"On Efficient Language and Vision Assistants for Visually-Situated Natural Language Understanding: What Matters in Reading and Reasoning","year":2024,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Korea Advanced Institute of Science and Technology","keywords":"Situated; Computer science; Reading (process); Natural language; Natural (archaeology); Natural language processing; Artificial intelligence; Human–computer interaction; Linguistics","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":[0.0003550253,0.0001223281,0.000113387,0.0002341271,0.0001090063,0.000826622,0.0001395264,0.00004195906,0.000004529855],"category_scores_gemma":[0.0000655603,0.0001019385,0.00002129399,0.0003095953,0.00002776246,0.0003038178,0.0001143854,0.000174464,0.000005745862],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001031878,"about_ca_system_score_gemma":0.00001171618,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001308782,"about_ca_topic_score_gemma":0.00003681436,"domain_scores_codex":[0.9990209,0.00004088154,0.0001359494,0.0004537117,0.0001427304,0.0002058596],"domain_scores_gemma":[0.9991886,0.000545439,0.00002908347,0.0001675972,0.000009307826,0.00005999389],"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":[0.0001230692,0.0001734466,0.001319806,0.000494433,0.000069021,0.0001564446,0.05507266,0.001971288,0.05112095,0.6398652,0.0007331326,0.2489006],"study_design_scores_gemma":[0.0003668944,0.00005899407,0.004467181,0.0006229634,0.000005181359,0.00001282938,0.003235783,0.9903799,0.000359362,0.0003255908,0.00001361091,0.0001516515],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7165072,0.0007109378,0.27956,0.002424579,0.0001262028,0.0002596573,0.000001569417,0.0002339785,0.0001758872],"genre_scores_gemma":[0.9902537,0.00000970871,0.009232372,0.0003418415,0.0000126555,0.00001764141,0.000007498581,0.00001425936,0.0001103444],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9884087,"threshold_uncertainty_score":0.7971135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01200450638272115,"score_gpt":0.3344200225874453,"score_spread":0.3224155162047241,"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."}}