{"id":"W3211759539","doi":"","title":"Referring Transformer: A One-step Approach to Multi-task Visual Grounding","year":2021,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Transformer; Artificial intelligence; Segmentation; Encoder; Ground; Leverage (statistics); Margin (machine learning); Language model; Machine learning; Natural language processing; Engineering","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.000435901,0.0002237934,0.0002671666,0.0002610426,0.0005347241,0.001660799,0.0006126375,0.0001032169,0.000002397775],"category_scores_gemma":[0.000106688,0.0002251534,0.00006738031,0.001136208,0.00001944586,0.003809819,0.00008281011,0.0003364967,0.0001409706],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001510142,"about_ca_system_score_gemma":0.000140661,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000290276,"about_ca_topic_score_gemma":0.000005803499,"domain_scores_codex":[0.9978309,0.0001025296,0.0007379532,0.0003678793,0.0005512963,0.0004094261],"domain_scores_gemma":[0.9987611,0.00004062635,0.0002356168,0.0004002431,0.0003659928,0.0001964078],"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":[0.00002786859,0.0005352687,0.001904583,0.00311326,0.00006795787,0.000005301476,0.03552091,0.1823048,0.01674969,0.03633466,0.0002651694,0.7231706],"study_design_scores_gemma":[0.0004010367,0.00002822821,0.002590489,0.0001235089,0.000007715937,0.00009470901,0.0006777712,0.9913728,0.0005695572,0.000007659619,0.003844104,0.0002823851],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04804065,0.00009971613,0.945763,0.0005466587,0.0002751439,0.0004973779,0.000003541508,0.0006078011,0.004166141],"genre_scores_gemma":[0.9345828,0.000001785505,0.06432245,0.0005670213,0.00008420798,0.0002418658,0.00005121257,0.00001559823,0.0001330518],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8865421,"threshold_uncertainty_score":0.9993756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04019012310903781,"score_gpt":0.3048863225568016,"score_spread":0.2646961994477638,"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."}}