{"id":"W4385570798","doi":"10.18653/v1/2023.acl-long.136","title":"Augmentation-Adapted Retriever Improves Generalization of Language Models as Generic Plug-In","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Tsinghua University","keywords":"Generalization; Computer science; Scheme (mathematics); Source code; Labrador Retriever; Open source; Plug-in; Set (abstract data type); Plug and play; Code (set theory); Artificial intelligence; Operating system; Mathematics; Programming language","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.001800025,0.002627676,0.001466196,0.0009594684,0.0004665918,0.001327807,0.002277973,0.001395849,0.008124816],"category_scores_gemma":[0.005851092,0.0006922616,0.001907303,0.000772331,0.0006542777,0.003847069,0.002583617,0.002405489,0.01202364],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005852274,"about_ca_system_score_gemma":0.0008410865,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004947495,"about_ca_topic_score_gemma":0.008355929,"domain_scores_codex":[0.9986368,0.0004114463,0.00008062786,0.0005132306,0.0002244127,0.0001335618],"domain_scores_gemma":[0.9984111,0.0005499672,0.00006311736,0.0006682318,0.0002417001,0.00006596167],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0009198122,0.0005849526,0.002220668,0.0007422203,0.0003436215,0.0007015707,0.000516113,0.1046455,0.0823286,0.003712394,0.05543081,0.7478539],"study_design_scores_gemma":[0.0001297415,0.0003664499,0.001135717,0.0000443867,0.0001414744,0.0004592547,0.0001891348,0.9268488,0.04613468,0.005847133,0.01860269,0.0001006028],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1014101,0.003563711,0.7171318,0.0007697109,0.0005357227,0.000352477,0.003596098,0.1608335,0.01180687],"genre_scores_gemma":[0.5265075,0.001504941,0.4128579,0.001877631,0.0003478818,0.0007190304,0.02033209,0.00824588,0.027607],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008124816,"threshold_uncertainty_score":0.02718025,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03114723848441501,"score_gpt":0.2720054915364587,"score_spread":0.2408582530520437,"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."}}