{"id":"W4386566659","doi":"10.18653/v1/2023.eacl-main.239","title":"DyLoRA: Parameter-Efficient Tuning of Pre-trained Models using Dynamic Search-Free Low-Rank Adaptation","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":94,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Benchmark (surveying); Rank (graph theory); Sorting; Language model; Range (aeronautics); Task (project management); Machine learning; Artificial intelligence; Algorithm","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":[0.0006165989,0.0001328196,0.0001868974,0.0002878802,0.00009614532,0.00008044231,0.0008930618,0.00006285454,0.000006116624],"category_scores_gemma":[0.00008301611,0.0001270223,0.00007560732,0.0008151265,0.00004187586,0.0003657474,0.0005029347,0.0001210762,0.000007548852],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000077994,"about_ca_system_score_gemma":0.0001151761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002611162,"about_ca_topic_score_gemma":0.00002550852,"domain_scores_codex":[0.9982311,0.00008308057,0.0003623722,0.0004259833,0.0005245568,0.0003728943],"domain_scores_gemma":[0.9986926,0.0001749783,0.00008364212,0.0008437643,0.0001224763,0.00008250347],"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":[0.000004533765,0.00002840835,0.000006099712,0.00003279478,0.000009255813,0.000004968706,0.002527373,0.9655178,0.005094908,0.01477012,0.000004334555,0.01199946],"study_design_scores_gemma":[0.0003589346,0.00003025371,0.0001192532,0.0000592977,0.000004903926,0.000003034238,0.0002357124,0.9905726,0.0007644402,0.007716843,7.795092e-7,0.0001339106],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4112985,0.00001205908,0.5879031,0.0001379294,0.0001119554,0.0001420977,0.000001934803,0.0002087498,0.000183584],"genre_scores_gemma":[0.7614658,0.000002783239,0.2383658,0.00002792306,0.00001132348,0.000005110841,0.000002871488,0.00001049742,0.0001079065],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.3501672,"threshold_uncertainty_score":0.5179818,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07253695507708174,"score_gpt":0.29376392995067,"score_spread":0.2212269748735883,"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."}}