{"id":"W4386566893","doi":"10.18653/v1/2023.eacl-main.49","title":"Combining Parameter-efficient Modules for Task-level Generalisation","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute; McGill University","funders":"Institut de Valorisation des Données; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Reinforcement learning; Modular design; Benchmark (surveying); Machine learning; Artificial intelligence; Task (project management); Latent variable; Language model","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003016638,0.00007107112,0.00008196894,0.00009274813,0.000101791,0.0001026558,0.0003579476,0.000031563,0.000003256661],"category_scores_gemma":[0.00006351425,0.00006416528,0.00004494188,0.0002259178,0.000009278527,0.0001110254,0.0001386822,0.00003325815,0.00004663311],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002241063,"about_ca_system_score_gemma":0.0000183927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003210756,"about_ca_topic_score_gemma":0.00000482227,"domain_scores_codex":[0.9991584,0.00001867296,0.0001585912,0.0002851417,0.0001587504,0.0002204607],"domain_scores_gemma":[0.9993932,0.000140923,0.00003610066,0.0003410757,0.00004656694,0.00004208075],"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":[0.000001759987,0.00002918196,0.0001263099,0.00001606488,0.00001099616,0.000001937792,0.001114228,0.3552157,0.003420322,0.5449061,0.002434759,0.09272262],"study_design_scores_gemma":[0.000181979,0.00001846558,0.0004875687,0.000004598428,0.000001483493,9.840738e-7,0.00002668299,0.9863331,0.001879762,0.01029337,0.0006825018,0.00008952851],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1622808,0.000006282585,0.8355967,0.0009006997,0.0003918677,0.0001471776,0.000002508439,0.0003584127,0.0003156403],"genre_scores_gemma":[0.7134708,0.000001015896,0.2855226,0.0002305258,0.00004303716,0.00003927591,0.00000697824,0.000005286959,0.0006803909],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6311173,"threshold_uncertainty_score":0.2616584,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1278035831734825,"score_gpt":0.2946614849743965,"score_spread":0.1668579018009139,"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."}}