{"id":"W3171517119","doi":"","title":"LTL2Action: Generalizing LTL Instructions for Multi-Task RL","year":2021,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"","keywords":"Computer science; Task (project management); Reinforcement learning; Syntax; Semantics (computer science); Overhead (engineering); Linear temporal logic; Artificial intelligence; Scheme (mathematics); Sample complexity; Programming language; Theoretical computer science; Natural language processing; Mathematics","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":[0.000170645,0.0001460795,0.0001272332,0.0001553566,0.0002572799,0.0004134678,0.0006266681,0.00007227651,0.0001174249],"category_scores_gemma":[0.0003779239,0.0001451566,0.00008422847,0.0001951525,0.00002518843,0.0004708887,0.0001969761,0.0004518776,0.00002216068],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009417656,"about_ca_system_score_gemma":0.0001139896,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006148399,"about_ca_topic_score_gemma":0.00004724401,"domain_scores_codex":[0.9988263,0.00007135933,0.0002095072,0.0004243002,0.0002845536,0.0001840265],"domain_scores_gemma":[0.9990024,0.00006583213,0.0001419649,0.0002175852,0.000516413,0.0000558181],"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":[0.00001602399,0.00007188713,0.000937093,0.0000165035,0.00005162595,0.00003509981,0.0004068987,0.0005517834,0.03667066,0.8294941,0.0001846627,0.1315637],"study_design_scores_gemma":[0.0006104093,0.00008192129,0.0002830084,0.0001142561,0.000007933647,0.0001452645,0.00005491958,0.9203968,0.02601055,0.02425036,0.02771687,0.0003276846],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002352834,0.0003220849,0.9891091,0.004422527,0.00105958,0.00009345877,0.0000108311,0.000527754,0.002101824],"genre_scores_gemma":[0.471664,0.00003510555,0.5251624,0.0004314953,0.0001607533,0.00003285662,0.00005968806,0.00001175647,0.002441927],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.919845,"threshold_uncertainty_score":0.5919316,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07083520501108485,"score_gpt":0.3596141055672882,"score_spread":0.2887789005562033,"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."}}