{"id":"W4221147223","doi":"10.18653/v1/2022.acl-long.359","title":"E-LANG: Energy-Based Joint Inferencing of Super and Swift Language Models","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Huawei Technologies (Canada)","funders":"","keywords":"Computer science; Swift; Inference; Language model; Computation; Code (set theory); Encoder; Set (abstract data type); Artificial intelligence; Sequence (biology); Architecture; Energy (signal processing); Computer engineering; Machine learning; Theoretical computer science; Programming language; Operating system","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":[0.001141607,0.0001368521,0.000268257,0.0001273445,0.0003515777,0.00004138656,0.000776483,0.00005525384,0.000001793502],"category_scores_gemma":[0.003340005,0.0001146631,0.0001781117,0.000367194,0.00004706829,0.00007773642,0.000636578,0.0001675195,8.720863e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001604579,"about_ca_system_score_gemma":0.0001706605,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001198661,"about_ca_topic_score_gemma":0.000006187104,"domain_scores_codex":[0.9980941,0.00003528852,0.0005559971,0.000259414,0.0008460017,0.0002092169],"domain_scores_gemma":[0.9968638,0.0004126834,0.001126933,0.0001455939,0.001411608,0.00003931381],"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":[0.0000265153,0.000100008,0.04257328,0.0002696259,0.0001015388,1.378015e-7,0.00520541,0.7123318,0.001696462,0.2369406,0.0003582331,0.0003964188],"study_design_scores_gemma":[0.0005570599,0.00007969834,0.004652092,0.0001239829,0.00006269144,0.00000127604,0.0007201133,0.9663158,0.002816884,0.02405202,0.0004477281,0.0001706612],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8806838,0.000523417,0.09596352,0.002886951,0.002792329,0.001352779,0.0006908777,0.000218635,0.01488773],"genre_scores_gemma":[0.9791315,8.033768e-7,0.02024756,0.0001811979,0.0001025198,0.00001771853,0.000006253466,0.00001492594,0.0002974811],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.253984,"threshold_uncertainty_score":0.4675828,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01135981089682939,"score_gpt":0.2152195700374592,"score_spread":0.2038597591406298,"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."}}