{"id":"W3207306342","doi":"10.18653/v1/2022.acl-long.448","title":"Compositional Generalization in Dependency Parsing","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; McGill University; Canadian Institute for Advanced Research","keywords":"Parsing; Computer science; Dependency grammar; Dependency (UML); Artificial intelligence; Natural language processing; Generalization; Test set; Set (abstract data type); Divergence (linguistics); Programming language; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009917457,0.001453659,0.001314931,0.002710076,0.00212796,0.002552318,0.001996306,0.002063026,0.004985412],"category_scores_gemma":[0.03498561,0.0008361242,0.002306662,0.002887451,0.002182274,0.006937485,0.00459037,0.003816513,0.002246692],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002225199,"about_ca_system_score_gemma":0.003050086,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00637715,"about_ca_topic_score_gemma":0.0115612,"domain_scores_codex":[0.9920821,0.003512535,0.0004646268,0.002235059,0.001269764,0.0004357844],"domain_scores_gemma":[0.9743824,0.01619304,0.0009189369,0.005464439,0.00265057,0.0003906309],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001219857,0.0004623905,0.03833993,0.001122033,0.0005583718,0.001116676,0.002946438,0.1793848,0.02497991,0.07404023,0.06070839,0.6151209],"study_design_scores_gemma":[0.0001285098,0.0002598404,0.0127248,0.0001154689,0.000172747,0.0009485262,0.0005897931,0.7279741,0.02165438,0.2087022,0.02657006,0.0001595338],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2563904,0.001482692,0.6935202,0.001908713,0.0003163583,0.0005601835,0.008253627,0.02120875,0.01635902],"genre_scores_gemma":[0.682481,0.0004560244,0.2886207,0.001003095,0.0001556576,0.0005918663,0.02031038,0.002748833,0.003632484],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009917457,"threshold_uncertainty_score":0.05244911,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006549197846466241,"score_gpt":0.2383371770636452,"score_spread":0.231787979217179,"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."}}