{"id":"W7092849503","doi":"10.5281/zenodo.17395552","title":"InferBoost: A Fast Microservice Architecture for Reducing Large Language Model Hallucinations","year":2025,"lang":"en","type":"preprint","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Software System Performance and Reliability","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Weyerhauser (Canada)","funders":"","keywords":"Inference; Language model; Modular design; Architecture; Microservices; Latency (audio); Reliability (semiconductor)","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","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001163558,0.0002893108,0.0003122002,0.0004250142,0.002169245,0.001330628,0.003274477,0.0002489162,0.000217275],"category_scores_gemma":[0.0005373623,0.0002921359,0.0001703601,0.0005753602,0.0000659746,0.0002735128,0.006810669,0.0007248903,0.0004153969],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002617899,"about_ca_system_score_gemma":0.00004833116,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000442132,"about_ca_topic_score_gemma":0.000002876566,"domain_scores_codex":[0.9974186,0.0002527674,0.000439202,0.0009706712,0.0003783507,0.0005403872],"domain_scores_gemma":[0.9969668,0.00006283207,0.0002171235,0.001573505,0.001030167,0.0001495588],"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":[0.000172953,0.00126141,0.00001566957,0.009355123,0.000479476,0.00002944498,0.0873782,0.169601,0.005609951,0.03652406,0.2105752,0.4789974],"study_design_scores_gemma":[0.001083601,0.0001447768,0.0001562013,0.0006537929,0.00004692567,0.00006441884,0.0004323534,0.5157073,0.0009133884,0.003457246,0.4766457,0.0006942759],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00505442,0.0002330196,0.9715163,0.001917644,0.0003806216,0.001478879,0.001474572,0.001741098,0.0162035],"genre_scores_gemma":[0.8794818,0.0002125063,0.09593967,0.001613902,0.0008284944,0.000004999578,0.01021566,0.002367826,0.00933513],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8755766,"threshold_uncertainty_score":0.9999531,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02049488519489529,"score_gpt":0.2666741671606477,"score_spread":0.2461792819657525,"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."}}