{"paper":{"title":"Maximizing LLM Utility While Managing LLM Fallibility","arxiv_id":"2611.17020","primary_cat":"cs.RO","submitted_at":"2026-11-12T16:09:49+00:00","version_compiled":"v1"},"notice":"Every quoted sentence is the authors' text from the arXiv listing abstract. Labels and probabilities are the outputs of classifier-1.13.0 on the question bank abstract-p1-20260923, exactly as returned. No sentence in this record was written by a language model.","schema":"pith.compiled-record/v0","status":"compiled","abstract":{"source":"arXiv listing abstract v1","sentences":[{"i":0,"p":1.0,"id":"s0","own":false,"role":"background","text":"Large language models (LLMs) can supply autonomous systems with plans, predictions, and judgments, but these contributions are sometimes wrong.","hedge":{"p":0.52,"label":"hedged_possibility"},"own_p":0.06,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":1.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.45,"id":"s1","own":false,"role":"contribution","text":"This review asks how a system can maximize an LLM's utility while managing its fallibility.","hedge":{"p":0.62,"label":"unclear"},"own_p":0.17,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.2,"method":0.0,"result":0.0,"background":0.35,"limitation":0.0,"future_work":0.0,"contribution":0.45,"interpretation":0.0}},{"i":2,"p":0.76,"id":"s2","own":true,"role":"contribution","text":"Our answer is to adapt the system's reliance on the LLM as evidence about the LLM's performance accumulates.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.58,"main_claim_p":0.86,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.24,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.76,"interpretation":0.0}},{"i":3,"p":0.97,"id":"s3","own":false,"role":"background","text":"Systems built from imperfect components, such as noisy sensors and faulty software, have long worked in this way, and the classical methods they use can be applied to LLM-based systems as well.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.05,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.09,"role_probabilities":{"gap":0.0,"method":0.02,"result":0.0,"background":0.97,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":4,"p":0.68,"id":"s4","own":false,"role":"method","text":"A system can change its reliance on an LLM in several ways, depending on the LLM's role and on the other components in the system.","hedge":{"p":0.66,"label":"asserted"},"own_p":0.14,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.68,"result":0.0,"background":0.09,"limitation":0.0,"future_work":0.0,"contribution":0.22,"interpretation":0.01}},{"i":5,"p":0.9,"id":"s5","own":false,"role":"method","text":"Each change needs an estimate of the LLM's reliability, and the LLM's role and the kind of contribution it makes determine what to estimate.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.19,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.9,"result":0.0,"background":0.02,"limitation":0.0,"future_work":0.0,"contribution":0.08,"interpretation":0.0}},{"i":6,"p":0.9,"id":"s6","own":false,"role":"method","text":"Classical methods from decision theory, estimation, and online learning then turn these estimates into reliance decisions.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.08,"main_claim_p":0.1,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.9,"result":0.0,"background":0.04,"limitation":0.0,"future_work":0.0,"contribution":0.06,"interpretation":0.0}},{"i":7,"p":0.58,"id":"s7","own":false,"role":"result","text":"These decisions can reduce how often LLM errors affect the system.","hedge":{"p":0.73,"label":"hedged_possibility"},"own_p":0.13,"main_claim_p":0.14,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":0.0,"method":0.07,"result":0.58,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.04,"interpretation":0.31}},{"i":8,"p":0.5,"id":"s8","own":false,"role":"limitation","text":"However, some errors matter even when they are rare.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.07,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.11,"role_probabilities":{"gap":0.05,"method":0.0,"result":0.01,"background":0.37,"limitation":0.5,"future_work":0.0,"contribution":0.01,"interpretation":0.06}},{"i":9,"p":0.72,"id":"s9","own":false,"role":"method","text":"An independently verified mechanism can prevent such errors from being executed.","hedge":{"p":0.87,"label":"asserted"},"own_p":0.09,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.72,"result":0.02,"background":0.11,"limitation":0.02,"future_work":0.01,"contribution":0.08,"interpretation":0.04}},{"i":10,"p":0.37,"id":"s10","own":false,"role":"interpretation","text":"Without one, adapting reliance using reliability estimates can still keep their risk within a bound.","hedge":{"p":0.65,"label":"hedged_possibility"},"own_p":0.18,"main_claim_p":0.13,"quantitative":false,"quantitative_p":0.39,"role_probabilities":{"gap":0.0,"method":0.13,"result":0.34,"background":0.01,"limitation":0.07,"future_work":0.0,"contribution":0.08,"interpretation":0.37}},{"i":11,"p":0.58,"id":"s11","own":false,"role":"result","text":"Studies on adapting reliance on LLMs are emerging but limited, and the reviewed comparisons suggest that adapting reliance can improve system performance.","hedge":{"p":0.97,"label":"hedged_possibility"},"own_p":0.26,"main_claim_p":0.43,"quantitative":false,"quantitative_p":0.31,"role_probabilities":{"gap":0.22,"method":0.0,"result":0.58,"background":0.04,"limitation":0.04,"future_work":0.0,"contribution":0.05,"interpretation":0.07}}]},"compiled":{"main_claim":{"p":0.51,"text":"Our answer is to adapt the system's reliance on the LLM as evidence about the LLM's performance accumulates.","location":"abstract s2","hedge_strength":{"p":1.0,"label":"asserted"}},"provenance":{"route":{"via":"TypeSafe API","model":"classifier-1.13.0","endpoint":"https://api.typesafe.ai/v1/systemone","provider":"TypeSafe"},"packing":"one request per paper","requests":1,"input_tokens":8636,"cost_usd_reported":0.0},"compiled_at":"2026-11-13","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.04,"states_limitation":0.69,"human_participants":0.04,"controlled_comparison":0.24,"randomized_assignment":0.02,"states_quantitative_result":0.07,"compares_with_prior_results":0.64,"mentions_code_or_data_availability":0.02},"evidential_basis":{"p":0.91,"value":"literature","probabilities":{"argument":0.09,"literature":0.91,"proof_or_derivation":0.0,"analysis_of_existing_data":0.0,"observation_or_experiment":0.0,"simulation_or_computation":0.0}},"contribution_type":{"p":1.0,"value":"review","probabilities":{"review":1.0,"method_or_tool":0.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":0.81,"text":"Our answer is to adapt the system's reliance on the LLM as evidence about the LLM's performance accumulates.","location":"abstract s2"}},"coverage":{"tier":"abstract","levels_run":["P abstract"],"not_examined":["full text","figures and tables","references and citing sentences","relations to other papers"],"sources_read":["arXiv listing abstract v1"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":12,"sentences_in_abstract":12},"measured_accuracy":{"url":"https://pith.science/accuracy","measured_on":"2026-10-02","method":"two independent model families labelled a fresh random sample blind; the figure is how often the published label matches their consensus","sentence_role_agreement":0.92,"main_claim_agreement":0.908,"typed_citation_agreement":0.933,"typed_citation_agreement_95_interval":[0.917,0.945],"review_comments_accurate":0.758,"review_comments_accurate_95_interval":[0.684,0.82],"review_caught_admitted_error_found_or_related":0.448,"review_comments_judged":{"judged_by":["gpt-5.6-luna","grok-4.7"],"judged_against":"Pith's own text copy of each paper: text extracted from the PDF, the first 90,000 characters","measured_on":"2026-10-04","accurate":113,"decided":149,"sampled_reviews":200,"split_by_lane":false,"by_lane":"https://pith.science/accuracy#by-lane"}}}