{"id":"W3034552719","doi":"10.18653/v1/2020.acl-main.591","title":"Exploiting Syntactic Structure for Better Language Modeling: A Syntactic Distance Approach","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Canadian Institute for Advanced Research; Université de Montréal","funders":"National Natural Science Foundation of China; Westlake University","keywords":"Treebank; Computer science; Perplexity; Parsing; Natural language processing; Artificial intelligence; Language model; Task (project management); Representation (politics); Parse tree; Syntactic structure; Syntax","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002120436,0.0005076593,0.0005966133,0.0001120673,0.0001507883,0.0006440951,0.002124499,0.0003311234,0.0000152683],"category_scores_gemma":[0.0001966042,0.0004731519,0.0002654126,0.0001608898,0.0000175054,0.00044188,0.001825405,0.001023891,0.000006926409],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001656308,"about_ca_system_score_gemma":0.0001744295,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001018872,"about_ca_topic_score_gemma":0.00001178004,"domain_scores_codex":[0.9966173,0.00007559756,0.0005712393,0.0016876,0.0004778457,0.0005704896],"domain_scores_gemma":[0.9976665,0.0001588452,0.0002695024,0.001603722,0.0001093844,0.0001920831],"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":[0.0001056943,0.0002851381,0.0003104903,0.01035482,0.0008830218,0.0001914877,0.04186142,0.5636278,0.005573853,0.2762976,0.001034696,0.09947397],"study_design_scores_gemma":[0.0001993522,0.00001413311,0.000002247776,0.0001117087,0.00004246252,0.00001527183,0.0003643146,0.9659885,0.000379087,0.03228155,0.00007398544,0.0005273402],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.019699,0.0002673368,0.9743423,0.002358348,0.0006708837,0.0009036502,0.00004005253,0.0006477278,0.00107072],"genre_scores_gemma":[0.5698332,0.000002930902,0.4287651,0.0007576153,0.0003887111,0.0001185566,0.00003612343,0.00003732274,0.00006048133],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5501341,"threshold_uncertainty_score":0.999772,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05290269024819936,"score_gpt":0.2758101971273488,"score_spread":0.2229075068791495,"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."}}