{"id":"W3202332982","doi":"10.1609/aaai.v36i10.21306","title":"Retrieve, Caption, Generate: Visual Grounding for Enhancing Commonsense in Text Generation Models","year":2022,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Closed captioning; Commonsense reasoning; Computer science; Fluency; Transformer; Generative grammar; Commonsense knowledge; Artificial intelligence; Natural language processing; Image (mathematics); Linguistics; Knowledge extraction; 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":[],"consensus_categories":[],"category_scores_codex":[0.0008532511,0.0001163597,0.0001312733,0.0002091499,0.0006229937,0.0001459374,0.0004100615,0.00003452653,0.00005378521],"category_scores_gemma":[0.00005410363,0.0001303946,0.00004363855,0.000686238,0.00001211684,0.000348241,0.0002733264,0.0002267443,0.00001169298],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002597319,"about_ca_system_score_gemma":0.00008780927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001217292,"about_ca_topic_score_gemma":0.000658186,"domain_scores_codex":[0.9985455,0.0001808528,0.0003267405,0.0004437015,0.0002471728,0.0002560253],"domain_scores_gemma":[0.9992664,0.0001615755,0.00009854927,0.0003471805,0.00007115614,0.00005516421],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001232625,0.0001482105,0.000542105,0.000008528631,0.000007991524,0.000002465134,0.001590126,0.6449586,0.08197217,0.2581816,0.001140876,0.01143498],"study_design_scores_gemma":[0.0003345815,0.00006162122,0.0006285849,0.000001770532,0.000002168072,0.00001146894,0.00007884265,0.9926867,0.002371053,0.003162232,0.0004995907,0.0001613387],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3198872,0.00002200787,0.6780618,0.001197904,0.000127404,0.0003252715,0.000002643833,0.0001269111,0.0002487862],"genre_scores_gemma":[0.9081731,0.000001784616,0.09046747,0.0004360988,0.00009331213,0.0004405472,0.00004477945,0.00001491319,0.0003280404],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5882859,"threshold_uncertainty_score":0.5317336,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04270963079109104,"score_gpt":0.3035675939309457,"score_spread":0.2608579631398547,"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."}}