{"id":"W4315630815","doi":"10.1145/3571243","title":"Type-Preserving, Dependence-Aware Guide Generation for Sound, Effective Amortized Probabilistic Inference","year":2023,"lang":"en","type":"article","venue":"Proceedings of the ACM on Programming Languages","topic":"Formal Methods in Verification","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Soundness; Inference; Computer science; Type inference; Probabilistic logic; Fuzz testing; Theoretical computer science; Programming language; Exploit; Machine learning; Artificial intelligence","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.001336099,0.0001665824,0.0001900082,0.00012972,0.0002105173,0.0002269243,0.002858767,0.00008717246,0.000002052216],"category_scores_gemma":[0.02095066,0.0001214203,0.0000908718,0.0009428581,0.0000778406,0.0004421181,0.0009894805,0.0001449409,0.00001104297],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007114804,"about_ca_system_score_gemma":0.00003940134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002047708,"about_ca_topic_score_gemma":0.000009556029,"domain_scores_codex":[0.9985141,0.00003135546,0.0003038632,0.0004360882,0.0003940121,0.000320552],"domain_scores_gemma":[0.9978761,0.0004152516,0.0003336703,0.0007403086,0.0005881007,0.00004657792],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001854916,0.000293418,0.01566207,0.001829233,0.0001627329,0.000002751346,0.01080126,0.0007342634,0.1379442,0.2051561,0.007382374,0.6198461],"study_design_scores_gemma":[0.0018269,0.001779233,0.02241844,0.0008092509,0.0001560105,0.00003037449,0.00242357,0.2762403,0.5404788,0.1481262,0.004444115,0.001266849],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9473338,0.0001017109,0.04563677,0.0009576688,0.0008579074,0.003576121,0.000007316595,0.0009713461,0.0005573552],"genre_scores_gemma":[0.6957716,0.000005972505,0.3031808,0.00004693268,0.0001055793,0.0006115363,0.000005144971,0.00001831564,0.0002541251],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6185792,"threshold_uncertainty_score":0.9872963,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05363607663135606,"score_gpt":0.3643441365370954,"score_spread":0.3107080599057393,"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."}}