LLM Reliability

LLM: Large Language Model

Ask about this topic

Overview

It asks not only whether a model gets an answer right, but whether it knows when it may be wrong, follows constraints, uses evidence faithfully, and behaves consistently across repeated or multi-turn use. Researchers test these properties with adversarial and perturbation benchmarks, repeated-run and cross-language evaluations, domain-specific tasks, and audits of LLM judges and tool-using agents. Common approaches include estimating uncertainty from logits, hidden representations, or response agreement; selective abstention and conformal calibration; retrieval and symbolic verification; and training methods that reward grounded answers, correction, or hesitation. The central tension is that benchmark accuracy and verbal confidence often fail to predict reliability on individual cases, long tasks, or shifted inputs, while stronger checks and abstention trade coverage, latency, and cost for safer decisions.

Latest papers

CardsList

802 papers116 papers in the last four weeks, up 18% on the four weeks before. 0.7% of all new papers.

Apr 7, 2026

The Oracle's Fingerprint: Correlated AI Forecasting Errors and the Limits of Bias Transmission

When large language models (LLMs) are consulted as forecasting tools, the independence of individual errors -- the foundation of collective intelligence -- may collapse. We test three conditions necessary for this "epistemic monoculture" to emerge. In Study 1, we show that GPT-4o, Claude, and Gemini exhibit highly correlated forecasting errors on 568 resolved binary prediction questions (mean pairwise error correlation r = 0.77, p < 0.001; r = 0.78 excluding likely-leaked questions), despite being developed independently by different organizations. In Study 2, we test whether this correlated bias has propagated into human crowd forecasts, using a within-question design that tracks community prediction shifts across the ChatGPT launch boundary (November 2022). We find that community forecasts move in the direction predicted by LLMs (r = 0.20, p = 0.007), but this shift is fully explained by rational updating toward ground truth. In Study 3, we examine whether the category-level pattern of human forecasting errors increasingly resembles the LLM bias fingerprint. We find the opposite: pre-ChatGPT human biases already strongly resembled the LLM pattern (r = 0.87), while post-ChatGPT the resemblance weakened (r = -0.28). Together, these findings reveal an epistemic monoculture that is built but not yet activated: three nominally independent AI systems share the same failure modes, amplifying precisely the biases humans already hold.

Theodor Spiro

Correlated Errors in Language ModelsLLM ReliabilityLanguage Model Bias EvaluationProbabilistic Forecasting
AskPDF
Apr 7, 2026

A Multi-Stage Validation Framework for Trustworthy Large-scale Clinical Information Extraction using Large Language Models

Large language models (LLMs) show promise for extracting clinically meaningful information from unstructured health records, yet their translation into real-world settings is constrained by the lack of scalable and trustworthy validation approaches. Conventional evaluation methods rely heavily on annotation-intensive reference standards or incomplete structured data, limiting feasibility at population scale. We propose a multi-stage validation framework for LLM-based clinical information extraction that enables rigorous assessment under weak supervision. The framework integrates prompt calibration, rule-based plausibility filtering, semantic grounding assessment, targeted confirmatory evaluation using an independent higher-capacity judge LLM, selective expert review, and external predictive validity analysis to quantify uncertainty and characterize error modes without exhaustive manual annotation. We applied this framework to extraction of substance use disorder (SUD) diagnoses across 11 substance categories from 919,783 clinical notes. Rule-based filtering and semantic grounding removed 14.59% of LLM-positive extractions that were unsupported, irrelevant, or structurally implausible. For high-uncertainty cases, the judge LLM's assessments showed substantial agreement with subject matter expert review (Gwet's AC1=0.80). Using judge-evaluated outputs as references, the primary LLM achieved an F1 score of 0.80 under relaxed matching criteria. LLM-extracted SUD diagnoses also predicted subsequent engagement in SUD specialty care more accurately than structured-data baselines (AUC=0.80). These findings demonstrate that scalable, trustworthy deployment of LLM-based clinical information extraction is feasible without annotation-intensive evaluation.

Maria Mahbub, Gregory M. Dams, Josh Arnold +9

Oak Ridge National Laboratory, Oak Ridge, TN, USA · Program Evaluation and Resource Center, Office of Mental Health and Office of Suicide Prevention, Department of Veterans Affairs, Menlo Park, CA, USA · Vanderbilt University Medical Center, Nashville, TN, USA +4

Clinical Information ExtractionClinical Language Model EvaluationLLM ReliabilityAutomated EvaluationLLM Grounding
AskPDF
Apr 7, 2026

The Model Agreed, But Didn't Learn: Diagnosing Surface Compliance in Large Language Models

Large Language Models (LLMs) internalize vast world knowledge as parametric memory, yet inevitably inherit the staleness and errors of their source corpora. Consequently, ensuring the reliability and malleability of these internal representations is imperative for trustworthy real-world deployment. Knowledge editing offers a pivotal paradigm for surgically modifying memory without retraining. However, while recent editors demonstrate high success rates on standard benchmarks, it remains questionable whether current evaluation frameworks that rely on assessing output under specific prompting conditions can reliably authenticate genuine memory modification. In this work, we introduce a simple diagnostic framework that subjects models to discriminative self-assessment under in-context learning (ICL) settings that better reflect real-world application environments, specifically designed to scrutinize the subtle behavioral nuances induced by memory modifications. This probing reveals a pervasive phenomenon of Surface Compliance, where editors achieve high benchmark scores by merely mimicking target outputs without structurally overwriting internal beliefs. Moreover, we find that recursive modifications accumulate representational residues, triggering cognitive instability and permanently diminishing the reversibility of the model's memory state. These insights underscore the risks of current editing paradigms and highlight the pivotal role of robust memory modification in building trustworthy, long-term sustainable LLM systems. Code is available at https://github.com/XiaojieGu/SA-MCQ.

Xiaojie Gu, Ziying Huang, Weicong Hong +3

Independent Researcher · Cornell Tech · The Ohio State University +1

Knowledge EditingLLM ReliabilityBelief Updating in LLMsLLM Evaluation
AskPDF
Apr 7, 2026

Swiss-Bench 003: Evaluating LLM Reliability and Adversarial Security for Swiss Regulatory Contexts

The deployment of large language models (LLMs) in Swiss financial and regulatory contexts demands empirical evidence of both production reliability and adversarial security, dimensions not jointly operationalized in existing Swiss-focused evaluation frameworks. This paper introduces Swiss-Bench 003 (SBP-003), extending the HAAS (Helvetic AI Assessment Score) from six to eight dimensions by adding D7 (Self-Graded Reliability Proxy) and D8 (Adversarial Security). I evaluate ten frontier models across 808 Swiss-specific items in four languages (German, French, Italian, English), comprising seven Swiss-adapted benchmarks (Swiss TruthfulQA, Swiss IFEval, Swiss SimpleQA, Swiss NIAH, Swiss PII-Scope, System Prompt Leakage, and Swiss German Comprehension) targeting FINMA Guidance 08/2024, the revised Federal Act on Data Protection (nDSG), and OWASP Top 10 for LLMs. Self-graded D7 scores (73-94%) exceed externally judged D8 security scores (20-61%) by a wide margin, though these dimensions use non-comparable scoring regimes. System prompt leakage resistance ranges from 24.8% to 88.2%, while PII extraction defense remains weak (14-42%) across all models. Qwen 3.5 Plus achieves the highest self-graded D7 score (94.4%), while GPT-oss 120B achieves the highest D8 score (60.7%) despite being the lowest-cost model evaluated. All evaluations are zero-shot under provider default settings; D7 is self-graded and does not constitute independently validated accuracy. I provide conceptual mapping tables relating benchmark dimensions to FINMA model validation requirements, nDSG data protection obligations, and OWASP LLM risk categories.

Fatih Uenal

Department of Computer Science, University of Colorado Boulder

LLM EvaluationLLM ReliabilityLanguage Model Safety EvaluationAdversarial Robustness of Language ModelsMultilingual Language Model EvaluationPrivacy Leakage in Language Models
AskPDF
Apr 7, 2026

Identifying Influential N-grams in Confidence Calibration via Regression Analysis

While large language models (LLMs) improve performance by explicit reasoning, their responses are often overconfident, even though they include linguistic expressions demonstrating uncertainty. In this work, we identify what linguistic expressions are related to confidence by applying the regression method. Specifically, we predict confidence of those linguistic expressions in the reasoning parts of LLMs as the dependent variables and analyze the relationship between a specific nn-gram and confidence. Across multiple models and QA benchmarks, we show that LLMs remain overconfident when reasoning is involved and attribute this behavior to specific linguistic information. Interestingly, several of the extracted expressions coincide with cue phrases intentionally inserted on test-time scaling to improve reasoning performance. Through our test on causality and verification that the extracted linguistic information truly affects confidence, we reveal that confidence calibration is possible by simply suppressing those overconfident expressions without drops in performance.

Shintaro Ozaki, Wataru Hashimoto, Hidetaka Kamigaito +2

Nara Institute of Science and Technology (NAIST) · The University of Tokyo

Language Model CalibrationConfidence Estimation in Language ModelsLLM ReliabilityConfidence EstimationCausal Interventions in Language Models
AskPDF
Apr 7, 2026

LLM Psychosis: A Theoretical and Diagnostic Framework for Reality-Boundary Failures in Large Language Models

The deployment of large language models (LLMs) as interactive agents has exposed a category of behavioral failure that prevailing terminology, principally hallucination, fails to adequately characterize. This paper introduces LLM Psychosis as a structured theoretical framework for pathological breakdowns in model cognition that exhibit functional resemblance to clinically recognized psychotic disorders. Five hallmark features define the framework: reality-boundary dissolution, persistence of injected false beliefs, logical incoherence under impossible constraints, self-model instability, and epistemic overconfidence. We argue these constitute a qualitatively distinct failure mode rather than a mere intensification of ordinary factual error. To operationalize the framework, we propose the LLM Cognitive Integrity Scale (LCIS), a five-axis diagnostic instrument organized around Environmental Reality Interface (ERI), Premise Arbitration Integrity (PAI), Logical Constraint Recognition (LCR), Self-Model Integrity (SMI), and Epistemic Calibration Integrity (ECI). We administer a targeted adversarial probe battery to ChatGPT 5 (GPT-5, OpenAI) and report empirical findings for each axis, documenting both intact-integrity baseline responses and the specific psychosis-like failure signatures elicited under adversarial escalation. Results support a three-tier severity taxonomy: Type I (Confabulatory), Type II (Delusional), and Type III (Dissociative). We further formalize the delusional gradient, a self-reinforcing dynamic in which correction pressure intensifies rather than resolves psychosis-like states, as the most consequential failure mode for deployed systems. Implications for safety evaluation, high-stakes deployment screening, and mechanistic interpretability research are discussed.

Ashutosh Raj

Founder & CEO, NeuraCare AI | IIT Ropar TBIF Incubatee, India

LLM ReliabilityLanguage Model Safety EvaluationHallucination in Language Models
AskPDF
Apr 7, 2026

Confidence Should Be Calibrated More Than One Turn Deep

Large Language Models (LLMs) are increasingly applied in high-stakes domains such as finance, healthcare, and education, where reliable multi-turn interactions with users are essential. However, existing work on confidence estimation and calibration, a major approach to building trustworthy LLM systems, largely focuses on single-turn settings and overlooks the risks and potential of multi-turn conversations. In this work, we introduce the task of multi-turn calibration to reframe calibration from a static property into a dynamic challenge central to reliable multi-turn conversation, where calibrating model confidence at each turn conditioned on the conversation history is required. We first reveal the risks of this setting: using Expected Calibration Error at turn T (ECE@T), a new metric that tracks calibration dynamics over turns, we show that user feedback (e.g., persuasion) can degrade multi-turn calibration. To address this, we propose MTCal, which minimises ECE@T via a surrogate calibration target, and further leverage calibrated confidence in ConfChat, a decoding strategy that improves both factuality and consistency of the model response in multi-turn interactions. Extensive experiments demonstrate that MT-Cal achieves outstanding and consistent performance in multi-turn calibration, and ConfChat preserves and even enhances model performance in multi-turn interactions. Our results mark multi-turn calibration as one missing link for scaling LLM calibration toward safe, reliable, and real-world use.

Zhaohan Zhang, Chengzhengxu Li, Xiaoming Liu +3

Queen Mary University of London · Xi’an Jiaotong University

Confidence Estimation in Language ModelsLanguage Model CalibrationLLM Reliability
AskPDF
Apr 7, 2026

LLM-as-Judge for Semantic Judging of Powerline Segmentation in UAV Inspection

The deployment of lightweight segmentation models on drones for autonomous power line inspection presents a critical challenge: maintaining reliable performance under real-world conditions that differ from training data. Although compact architectures such as U-Net enable real-time onboard inference, their segmentation outputs can degrade unpredictably in adverse environments, raising safety concerns. In this work, we study the feasibility of using a large language model (LLM) as a semantic judge to assess the reliability of power line segmentation results produced by drone-mounted models. Rather than introducing a new inspection system, we formalize a watchdog scenario in which an offboard LLM evaluates segmentation overlays and examine whether such a judge can be trusted to behave consistently and perceptually coherently. To this end, we design two evaluation protocols that analyze the judge's repeatability and sensitivity. First, we assess repeatability by repeatedly querying the LLM with identical inputs and fixed prompts, measuring the stability of its quality scores and confidence estimates. Second, we evaluate perceptual sensitivity by introducing controlled visual corruptions (fog, rain, snow, shadow, and sunflare) and analyzing how the judge's outputs respond to progressive degradation in segmentation quality. Our results show that the LLM produces highly consistent categorical judgments under identical conditions while exhibiting appropriate declines in confidence as visual reliability deteriorates. Moreover, the judge remains responsive to perceptual cues such as missing or misidentified power lines, even under challenging conditions. These findings suggest that, when carefully constrained, an LLM can serve as a reliable semantic judge for monitoring segmentation quality in safety-critical aerial inspection tasks.

Akram Hossain, Rabab Abdelfattah, Xiaofeng Wang +1

School of Computing Sciences and Computer Engineering, The University of Southern Mississippi, Hattiesburg, MS 39406 USA. · Electrical Engineering Department, University of South Carolina, Columbia, SC, USA. · Computer Science Department, Faculty of Computers & Artificial Intelligence, Fayoum University, Fayoum, Egypt.

LLM ReliabilityVision-Language Model-as-a-JudgeIndustrial InspectionSemantic Segmentation
AskPDF
Apr 6, 2026

Justified or Just Convincing? Error Verifiability as a Dimension of LLM Quality

As LLMs are deployed in high-stakes settings, users must judge the correctness of individual responses, often relying on model-generated justifications such as reasoning chains or explanations. Yet, no standard measure exists for whether these justifications help users distinguish correct answers from incorrect ones. We formalize this idea as error verifiability and propose vbalv_{\text{bal}}, a balanced metric that measures whether justifications enable raters to accurately assess answer correctness, validated against human raters who show high agreement. We find that neither common approaches, such as post-training and model scaling, nor more targeted interventions recommended improve verifiability. We introduce two methods that succeed at improving verifiability: reflect-and-rephrase (RR) for mathematical reasoning and oracle-rephrase (OR) for factual QA, both of which improve verifiability by incorporating domain-appropriate external information. Together, our results establish error verifiability as a distinct dimension of response quality that does not emerge from accuracy improvements alone and requires dedicated, domain-aware methods to address.

Xiaoyuan Zhu, Kimberly Le Truong, Riccardo Fogliato +8

University of Southern California · Carnegie Mellon University · Microsoft Core AI +1

LLM EvaluationLLM Answer VerificationLLM Reliability
AskPDF
Apr 6, 2026

Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction

Large language models (LLMs) achieve high accuracy in medical diagnosis when all clinical information is provided in a single turn, yet how they behave under multi-turn evidence accumulation closer to real clinical reasoning remains unexplored. We introduce MINT (Medical Incremental N-Turn Benchmark), a high-fidelity, multi-turn medical diagnosis benchmark comprising 1,035 cases with clinically labeled evidence shards, controlled turn granularity, and information-preserving decomposition. Through systematic evaluation of 11 LLMs on MINT, we uncover three persistent behavioral patterns that significantly impact diagnostic decisions: (1) intent to answer, models rush to answer before sufficient evidence has been observed, with over 55% of answers committed within the first two turns; (2) self-correction, incorrect-to-correct answer revisions occur at up to 10.6 times the rate of correct-to-incorrect flips, revealing a latent capacity for self-correction that premature commitment forecloses; and (3) strong lures, clinically salient information such as laboratory results trigger premature answering even when models are explicitly instructed to wait. We translate these findings into clinically actionable guidance: deferring the diagnostic question to later turns reduces premature answering and improves accuracy at the first point of commitment by up to 62.6%, while reserving salient clinical evidence for later turns prevents a catastrophic accuracy drop of up to 23.3% caused by premature commitment. Our work provides both a controlled evaluation framework and concrete recommendations for improving the reliability of LLMs in multi-turn medical diagnosis.

Jinrui Fang, Runhan Chen, Xu Yang +9

The University of Texas at Austin · New York University · The University of Texas Southwestern Medical Center +2

Clinical Language Model EvaluationMedical DiagnosisLLM ReliabilityClinical ReasoningLLM Self-Correction
AskPDF
Apr 5, 2026

NeuroSymbolic AI for Legal AI-TRISM: Trustworthy, Reliable, Interpretable, Safe Models

Large Language Models (LLMs) have transformed natural language processing, but their lack of interpretable reasoning and tendency to hallucinate pose significant challenges for legal applications. While LLMs show promise for legal text analysis and generation, they struggle with accurate citation attribution and precedent verification. For example, in legal contexts, a single incorrect precedent can jeopardize a case. Current approaches to improve LLM reliability in legal domains suffer from two key limitations: inadequate integration of structured legal knowledge during training or fine-tuning, and insufficient verification mechanisms for generated legal content. To address these challenges, we propose the TRISM (Trustworthy, Reliable, Interpretable, Safe Models) framework, which integrates NeuroSymbolic AI principles with LLMs to leverage both neural learning capabilities and symbolic reasoning over structured legal knowledge. The TRISM approach addresses the above limitations while maintaining interpretable decision pathways. Our framework formalizes the extraction of symbolic knowledge from legal textual documents and incorporates Retrieval-Augmented Generation (RAG) as a core component for grounding LLM outputs in verified legal sources. In this position paper, we make the following contributions: (1) An analysis of the limitations of AI in law; (2) Introduce RASOR RAG which creates foundations for neurosymbolic RAG by generating explicit interpretable rationales that could be formalized into symbolic representations; (3) A formalized methodology for creating symbolic legal knowledge bases that support both interpretable reasoning and output verification in LLMs; and (4) The TRISM framework for integrating symbolic legal knowledge with LLMs.

Deepa Tilwani, Yash Saxena, Ankur Padia +2

Department of Computer Science, AI Institute, University of South Carolina, Columbia, South Carolina, USA · Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County, Baltimore, Maryland, USA · Department of Computer Science and Engineering, The Ohio State University, Columbus, Ohio, USA

LLM ReliabilityNeuro-Symbolic AILegal NLPLLM InterpretabilityLLM GroundingRetrieval-Augmented Generation
AskPDF
Apr 5, 2026

A Reproducible Semantic Benchmark for Multivendor DSM-to-CLI Translation

Translating high-level network intents into correct multivendor configurations remains a central challenge in network automation, as syntactically valid outputs may still violate the intended operational state. Despite recent advances in Large Language Models (LLMs), the field still lacks reproducible semantic benchmarks for rigorous cross-vendor evaluation. This paper presents a reproducible DSM-to-CLI semantic benchmark covering five cloud LLMs, three vendors, five representative use cases, and ten repeated runs per experimental cell under fixed judges and an explicit failure taxonomy. Our results show that semantic quality and operational reliability are orthogonal, vendor effects dominate use-case effects, and repeated-run dispersion strongly predicts vote instability, with Huawei VRP exposing failure modes hidden by aggregate metrics. These findings demonstrate that multivendor, repeated-execution semantic benchmarks are essential for scientifically rigorous comparison of LLM-based network configuration systems.

Jerônimo Menezes, Leonardo Bitzki, Diego Kreutz +3

Federal University of Rio Grande do Sul (UFRGS) · AI Horizon Labs and PPGES – Federal University of Pampa (UNIPAMPA)

LLM EvaluationLLM Reliability
AskPDF
Apr 4, 2026

Structural Rigidity and the 57-Token Predictive Window: A Physical Framework for Inference-Layer Governability in Large Language Models

Current AI safety relies on behavioral monitoring and post-training alignment, yet empirical measurement shows these approaches produce no detectable pre-commitment signal in a majority of instruction-tuned models tested. We present an energy-based governance framework connecting transformer inference dynamics to constraint-satisfaction models of neural computation, and apply it to a seven-model cohort across five geometric regimes. Using trajectory tension (rho = ||a|| / ||v||), we identify a 57-token pre-commitment window in Phi-3-mini-4k-instruct under greedy decoding on arithmetic constraint probes. This result is model-specific, task-specific, and configuration-specific, demonstrating that pre-commitment signals can exist but are not universal. We introduce a five-regime taxonomy of inference behavior: Authority Band, Late Signal, Inverted, Flat, and Scaffold-Selective. Energy asymmetry (Σ\r{ho}_misaligned / Σ\r{ho}_aligned) serves as a unifying metric of structural rigidity across these regimes. Across seven models, only one configuration exhibits a predictive signal prior to commitment; all others show silent failure, late detection, inverted dynamics, or flat geometry. We further demonstrate that factual hallucination produces no predictive signal across 72 test conditions, consistent with spurious attractor settling in the absence of a trained world-model constraint. These results establish that rule violation and hallucination are distinct failure modes with different detection requirements. Internal geometry monitoring is effective only where resistance exists; detection of factual confabulation requires external verification mechanisms. This work provides a measurable framework for inference-layer governability and introduces a taxonomy for evaluating deployment risk in autonomous AI systems.

Gregory M. Ruddell

SnailSafe.ai — Carson City, Nevada

Language Model Safety EvaluationInference-Time Language Model AlignmentLLM ReliabilityLLM Auditing
AskPDF
Apr 3, 2026

Measuring LLM Trust Allocation Across Conflicting Software Artifacts

LLM-based software engineering assistants often reason over multiple artifacts, including code, documentation, signatures, and tests, even when those artifacts are incomplete or mutually inconsistent. Existing evaluations primarily measure final outputs, leaving unclear whether a model recognized unreliable evidence, identified the faulty source, or prioritized the appropriate artifact. We introduce TRACE, a controlled method for evaluating how LLMs assess and prioritize conflicting software artifacts. TRACE constructs paired clean and perturbed versions of real-world Java method bundles by injecting known faults into the documentation, implementation, or both while holding the remaining artifacts fixed. Models then assess artifact quality, detect and localize inconsistencies, and rank the available sources by reliability. Using 22,339 valid responses from seven LLMs on 456 method bundles, we find that quality penalties are generally localized to the perturbed artifact and increase with fault severity. However, models exhibit a consistent source-origin asymmetry: they detect documentation faults at 67-94% and explicit documentation-implementation contradictions at 50-91%, but detection falls by 21-43 percentage points when only the implementation changes while documentation remains intact. Models also struggle to deprioritize faulty implementations, and confidence provides little separation between correct and incorrect judgments for six of seven models. These results show that current LLMs are not symmetric integrators of software evidence: they audit natural-language specifications more reliably than subtle implementation behavior. TRACE provides a controlled method for exposing this failure mode before LLMs are used in correctness-critical software engineering workflows.

Noshin Ulfat, Ahsanul Ameen Sabit, Soneya Binta Hossain

University of Texas at Dallas Texas, USA · University of Texas at Dallas Richardson, Texas, USA

LLM ReliabilitySoftware EngineeringLLM Auditing
AskPDF
Apr 3, 2026

BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence

Large language models (LLMs) often produce confident but incorrect answers in settings where abstention would be safer. Standard evaluation protocols, however, require a response and do not account for how confidence should guide decisions under different risk preferences. To address this gap, we introduce the Behavioral Alignment Score (BAS), a decision-theoretic metric for evaluating how well LLM confidence supports abstention-aware decision making. BAS is derived from an explicit answer-or-abstain utility model and aggregates realized utility across a continuum of risk thresholds, yielding a measure of decision-level reliability that depends on both the magnitude and ordering of confidence. We show theoretically that truthful confidence estimates uniquely maximize expected BAS utility, linking calibration to decision-optimal behavior. BAS is related to proper scoring rules such as log loss, but differs structurally: log loss penalizes underconfidence and overconfidence symmetrically, whereas BAS imposes an asymmetric penalty that strongly prioritizes avoiding overconfident errors. Using BAS alongside widely used metrics such as ECE and AURC, we then construct a benchmark of self-reported confidence reliability across multiple LLMs and tasks. Our results reveal substantial variation in decision-useful confidence, and while larger and more accurate models tend to achieve higher BAS, even frontier models remain prone to severe overconfidence. Importantly, models with similar ECE or AURC can exhibit very different BAS due to highly overconfident errors, highlighting limitations of standard metrics. We further show that simple interventions, such as top-kk confidence elicitation and post-hoc calibration, can meaningfully improve confidence reliability. Overall, our work provides both a principled metric and a comprehensive benchmark for evaluating LLM confidence reliability.

Sean Wu, Fredrik K. Gustafsson, Edward Phillips +3

Department of Engineering Science, University of Oxford · Oxford Suzhou Centre for Advanced Research

LLM EvaluationConfidence Estimation in Language ModelsLLM ReliabilityLanguage Model CalibrationLLM Abstention
AskPDF
Apr 3, 2026

BibTeX Citation Errors in Scientific Publishing Agents: Evaluation and Mitigation

Large language models with web search are increasingly used in scientific publishing agents, yet they produce BibTeX entries with pervasive field-level errors stemming from omission, partial corruption, substitution, and hallucination. We construct a benchmark of 931 papers across four domains and three citation tiers---popular, low-citation, and recent post-cutoff---with version-aware ground truth. Three search-enabled frontier models (GPT-5, Claude Sonnet-4.6, Gemini-3 Flash) generate approximately 23,000 field-level observations. Overall accuracy is 83.6%, but only 50.9% of entries are fully correct; accuracy drops 27.7 pp from popular to recent papers, revealing heavy reliance on parametric memory even when search is available. Co-occurrence analysis identifies two failure modes: wholesale entry substitution and isolated field error. We present clibib, an open-source tool for deterministic BibTeX retrieval, as a mitigation mechanism. Two-stage integration raises accuracy to 91.5% (+8.0 pp) and fully correct entries to 78.3%, with a 0.8% regression rate. Separating search from revision yields larger gains and lower regression than single-stage tool loops (0.8% vs. 4.8%), demonstrating that integration architecture matters independently of model/tool capability. We release clibib and an accompanying agent skill under the MIT License to improve citation accuracy in increasingly automated scientific workflows.

Delip Rao, Chris Callison-Burch

University of Pennsylvania

Citation VerificationLLM ReliabilityScientific Workflow AutomationLanguage Model Error Detection
AskPDF
Apr 2, 2026

Measuring What Cannot Be Surveyed: LLMs as Instruments for Latent Cognitive Variables in Labor Economics

This paper establishes the theoretical and practical foundations for using Large Language Models (LLMs) as measurement instruments for latent economic variables -- specifically variables that describe the cognitive content of occupational tasks at a level of granularity not achievable with existing survey instruments. I formalize four conditions under which LLM-generated scores constitute valid instruments: semantic exogeneity, construct relevance, monotonicity, and model invariance. I then apply this framework to the Augmented Human Capital Index (AHC_o), constructed from 18,796 O*NET task statements scored by Claude Haiku 4.5, and validated against six existing AI exposure indices. The index shows strong convergent validity (r = 0.85 with Eloundou GPT-gamma, r = 0.79 with Felten AIOE) and discriminant validity. Principal component analysis confirms that AI-related occupational measures span two distinct dimensions -- augmentation and substitution. Inter-rater reliability across two LLM models (n = 3,666 paired scores) yields Pearson r = 0.76 and Krippendorff's alpha = 0.71. Prompt sensitivity analysis across four alternative framings shows that task-level rankings are robust. Obviously Related Instrumental Variables (ORIV) estimation recovers coefficients 25% larger than OLS, consistent with classical measurement error attenuation. The methodology generalizes beyond labor economics to any domain where semantic content must be quantified at scale.

Cristian Espinal Maya

Department of Economics, Universidad EAFIT, Medellín, Colombia.

AI and LaborLLM Reliability
AskPDF
Apr 2, 2026

Bayesian Elicitation with LLMs: Model Size Helps, Extra "Reasoning" Doesn't Always

Large language models (LLMs) have been proposed as alternatives to human experts for estimating unknown quantities with associated uncertainty, a process known as Bayesian elicitation. We test this by asking eleven LLMs to estimate population statistics, such as health prevalence rates, personality trait distributions, and labor market figures, and to express their uncertainty as 95% credible intervals. We vary each model's reasoning effort (low, medium, high) to test whether more "thinking" improves results. Our findings reveal three key results. First, larger, more capable models produce more accurate estimates, but increasing reasoning effort provides no consistent benefit. Second, all models are severely overconfident: their 95% intervals contain the true value only 9--44% of the time, far below the expected 95%. Third, a statistical recalibration technique called conformal prediction can correct this overconfidence, expanding the intervals to achieve the intended coverage. In a preliminary experiment, giving models web search access degraded predictions for already-accurate models, while modestly improving predictions for weaker ones. Models performed well on commonly discussed topics but struggled with specialized health data. These results indicate that LLM uncertainty estimates require statistical correction before they can be used in decision-making.

Luka Hobor, Mario Brcic, Mihael Kovac +1

University of Zagreb Faculty of Electrical Engineering and Computing, Zagreb, Croatia

Probability Elicitation from Language ModelsLLM Uncertainty EstimationLLM ReliabilityConfidence Estimation
AskPDF
Apr 2, 2026

Hidden Reliability Risks in Large Language Models: Systematic Identification of Precision-Induced Output Disagreements

Large language models (LLMs) are increasingly deployed under diverse numerical precision configurations, including standard floating-point formats (e.g., bfloat16 and float16) and quantized integer formats (e.g., int16 and int8), to meet efficiency and resource constraints. However, minor inconsistencies between LLMs of different precisions are difficult to detect and are often overlooked by existing evaluation methods. In this paper, we present PrecisionDiff, an automated differential testing framework for systematically detecting precision-induced behavioral disagreements in LLMs. PrecisionDiff generates precision-sensitive test inputs and performs cross-precision comparative analysis to uncover subtle divergences that remain hidden under conventional testing strategies. To demonstrate its practical significance, we instantiate PrecisionDiff on the alignment verification task, where precision-induced disagreements manifest as jailbreak divergence-inputs that are rejected under one precision may produce harmful responses under another. Experimental results show that such behavioral disagreements are widespread across multiple open-source aligned LLMs and precision settings, and that PrecisionDiff significantly outperforms vanilla testing methods in detecting these issues. Our work enables automated precision-sensitive test generation, facilitating effective pre-deployment evaluation and improving precision robustness during training.

Yifei Wang, Tianlin Li, Xiaohan Zhang +4

Shanghai Jiao Tong University, China · Beihang University, China · Nanyang Technological University, Singapore +1

LLM ReliabilityLLM EvaluationLanguage Model Safety Evaluation
AskPDF
Apr 2, 2026

What Is Actually Being Annotated? Inter-Prompt Reliability as a Measurement Problem in LLM-Based Social Science Labeling

Large language models (LLMs) are increasingly used for annotation in computational social science, yet their methodological reliability under prompt variation remains unclear. This paper introduces Inter-Prompt Reliability (IPR), a framework for evaluating the stability of LLM outputs across semantically equivalent but linguistically varied prompts. Drawing on Inter-Rater Reliability, IPR is measured by Pairwise Agreement Rate (PAR) and its distribution to capture both consistency and stochasticity in model behavior. We evaluate this framework on two tasks with distinct properties: TREC (interpretative) and Politifact (knowledge-anchored). Results show that LLM annotation exhibits substantial stochastic variation in interpretative tasks, while appearing more stable in knowledge-based tasks. We further show that majority voting across prompts significantly improves reproducibility and reduces variance. These findings suggest that LLM prompt acts as an instrumental measurement while its wording exhibits methodological uncertainty. For future LLM-based CSS studies, we suggest that researchers move beyond single-prompt evaluation toward distributional stability and prompt aggregation within our IPR framework.

Jingyuan Liu

Boston University

LLM-Assisted AnnotationLLM ReliabilityComputational Social ScienceInter-Rater Reliability
AskPDF
Apr 1, 2026

Generative AI-Based Virtual Assistant using Retrieval-Augmented Generation: An evaluation study for bachelor projects

Large Language Models have been increasingly employed in the creation of Virtual Assistants due to their ability to generate human-like text and handle complex inquiries. While these models hold great promise, challenges such as hallucinations, missing information, and the difficulty of providing accurate and context-specific responses persist, particularly when applied to highly specialized content domains. In this paper, we focus on addressing these challenges by developing a virtual assistant designed to support students at Maastricht University in navigating project-specific regulations. We propose a virtual assistant based on a Retrieval-Augmented Generation system that enhances the accuracy and reliability of responses by integrating up-to-date, domain-specific knowledge. Through a robust evaluation framework and real-life testing, we demonstrate that our virtual assistant can effectively meet the needs of students while addressing the inherent challenges of applying Large Language Models to a specialized educational context. This work contributes to the ongoing discourse on improving LLM-based systems for specific applications and highlights areas for further research.

Dumitru Verşebeniuc, Martijn Elands, Sara Falahatkar +4

Department of Advanced Computing Sciences, Maastricht University, Maastricht 6200 MD, The Netherlands

Educational QAQuestion AnsweringLLM ReliabilityRetrieval-Augmented GenerationLLM Evaluation
AskPDF
Apr 1, 2026

Exploring Silent Data Corruption as a Reliability Challenge in LLM Training

As Large Language Models (LLMs) scale in size and complexity, the consequences of failures during training become increasingly severe. A major challenge arises from Silent Data Corruption (SDC): hardware-induced faults that bypass system-level detection mechanisms. SDC may behave like benign numerical noise, but can also cause harmful gradient corruption that leads to loss spikes, divergence, or stalled progress. This work provides a controlled study of how intermittent SDC affects LLM pretraining. Using targeted fault injection at the level of GPU matrix-multiply instructions, we characterize the sensitivity of different bit positions, kernel functions, and execution stages. Our analysis shows that locally originating faults can produce impactful corruption, including NaN propagation, short-lived spikes in loss, gradient norm, and attention logits, as well as persistent parameter divergence. Building on the observed corruption signatures, we propose a lightweight detection method that identifies potentially harmful parameter updates. Experiments on LLaMA models with 60M, 350M, and 1.3B parameters demonstrate that recomputing the most recent training step upon detection can effectively mitigate the impact of these events.

Anton Altenbernd, Philipp Wiesner, Odej Kao

Technische Universität Berlin, Berlin, Germany

LLM ReliabilityLLM TrainingFault Detection
AskPDF
Apr 1, 2026

Towards Reliable Truth-Aligned Uncertainty Estimation in Large Language Models

Uncertainty estimation (UE) aims to quantify uncertainty in large language model (LLM) outputs to flag potential hallucinations and support reliable deployment. However, UE metrics often exhibit unstable performance across models and datasets, limiting their practical utility. We formalise this phenomenon as proxy failure, arising because UE metrics are derived from model behaviour rather than being explicitly grounded in factual correctness. We show that when mutual information between a UE score and correctness is low, the score's discriminability must be close to chance. To address this, we propose Truth AnChoring (TAC), a lightweight post-hoc calibration method that maps raw UE scores to truth-aligned correctness probabilities. Even with noisy and few-shot supervision, TAC continues to improve calibration, providing a practical calibration protocol. Our findings highlight the limitations of treating heuristic UE metrics as direct indicators of uncertainty about factual correctness, and position our TAC as a principled step toward more reliable uncertainty estimation for LLMs. The code repository is available at https://github.com/ponhvoan/TruthAnchor/.

Ponhvoan Srey, Quang Minh Nguyen, Xiaobao Wu +1

Nanyang Technological University, Singapore · KAIST, South Korea · Shanghai Jiao Tong University

LLM Uncertainty EstimationUncertainty CalibrationLLM ReliabilityLanguage Model Calibration
AskPDF
Apr 1, 2026

Decision-Centric Design for LLM Systems

LLM systems must make control decisions in addition to generating outputs: whether to answer, clarify, retrieve, call tools, repair, or escalate. In many current architectures, these decisions remain implicit within generation, entangling assessment and action in a single model call and making failures hard to inspect, constrain, or repair. We propose a decision-centric framework that separates decision-relevant signals from the policy that maps them to actions, turning control into an explicit and inspectable layer of the system. This separation supports attribution of failures to signal estimation, decision policy, or execution, and enables modular improvement of each component. It unifies familiar single-step settings such as routing and adaptive inference, and extends naturally to sequential settings in which actions alter the information available before acting. Across three controlled experiments, the framework reduces futile actions, improves task success, and reveals interpretable failure modes. More broadly, it offers a general architectural principle for building more reliable, controllable, and diagnosable LLM systems.

Wei Sun

IBM Research

LLM Decision-MakingLanguage Model-Based ControlLLM Reliability
AskPDF
Mar 31, 2026

Can Large Language Models Self-Correct in Medical Question Answering? An Exploratory Study

Large language models (LLMs) have achieved strong performance on medical question answering (medical QA), and chain-of-thought (CoT) prompting has further improved results by eliciting explicit intermediate reasoning; meanwhile, self-reflective (self-corrective) prompting has been widely claimed to enhance model reliability by prompting LLMs to critique and revise their own reasoning, yet its effectiveness in safety-critical medical settings remains unclear. In this work, we conduct an exploratory analysis of self-reflective reasoning for medical multiple-choice question answering: using GPT-4o and GPT-4o-mini, we compare standard CoT prompting with an iterative self-reflection loop and track how predictions evolve across reflection steps on three widely used medical QA benchmarks (MedQA, HeadQA, and PubMedQA). We analyze whether self-reflection leads to error correction, error persistence, or the introduction of new errors. Our results show that self-reflective prompting does not consistently improve accuracy and its impact is highly dataset- and model-dependent: it yields modest gains on MedQA but provides limited or negative benefits on HeadQA and PubMedQA, and increasing the number of reflection steps does not guarantee better performance. These findings highlight a gap between reasoning transparency and reasoning correctness, suggesting that self-reflective reasoning is better viewed as an analytical tool for understanding model behavior rather than a standalone solution for improving medical QA reliability.

Zaifu Zhan, Mengyuan Cui, Rui Zhang

Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, USA · Division of Computational Health Sciences, Department of Surgery, University of Minnesota, Minneapolis, USA · Department of Software Engineering, University of St. Thomas, Saint Paul,USA

LLM Self-CorrectionLLM ReliabilityMedical QAClinical Language Model Evaluation
AskPDF
Mar 31, 2026

Learning to Generate Formally Verifiable Step-by-Step Logic Reasoning via Structured Formal Intermediaries

Large language models (LLMs) have recently demonstrated impressive performance on complex, multi-step reasoning tasks, especially when post-trained with outcome-rewarded reinforcement learning Guo et al. 2025. However, it has been observed that outcome rewards often overlook flawed intermediate steps, leading to unreliable reasoning steps even when final answers are correct. To address this unreliable reasoning, we propose PRoSFI (Process Reward over Structured Formal Intermediates), a novel reward method that enhances reasoning reliability without compromising accuracy. Instead of generating formal proofs directly, which is rarely accomplishable for a modest-sized (7B) model, the model outputs structured intermediate steps aligned with its natural language reasoning. Each step is then verified by a formal prover. Only fully validated reasoning chains receive high rewards. The integration of formal verification guides the model towards generating step-by-step machine-checkable proofs, thereby yielding more credible final answers. PRoSFI offers a simple and effective approach to training trustworthy reasoning models.

Luoxin Chen, Yichi Zhou, Huishuai Zhang

Wangxuan Institue of Computer Technology, Peking University. · ByteDance.

LLM ReasoningLLM ReliabilityReinforcement Learning with Verifiable RewardsProcess Reward Models
AskPDF
Mar 31, 2026

Security in LLM-as-a-Judge: A Comprehensive SoK

LLM-as-a-Judge (LaaJ) is a novel paradigm in which powerful language models are used to assess the quality, safety, or correctness of generated outputs. While this paradigm has significantly improved the scalability and efficiency of evaluation processes, it also introduces novel security risks and reliability concerns that remain largely unexplored. In particular, LLM-based judges can become both targets of adversarial manipulation and instruments through which attacks are conducted, potentially compromising the trustworthiness of evaluation pipelines. In this paper, we present the first Systematization of Knowledge (SoK) focusing on the security aspects of LLM-as-a-Judge systems. We perform a comprehensive literature review across major academic databases, analyzing 863 works and selecting 45 relevant studies published between 2020 and 2026. Based on this study, we propose a taxonomy that organizes recent research according to the role played by LLM-as-a-Judge in the security landscape, distinguishing between attacks targeting LaaJ systems, attacks performed through LaaJ, defenses leveraging LaaJ for security purposes, and applications where LaaJ is used as an evaluation strategy in security-related domains. We further provide a comparative analysis of existing approaches, highlighting current limitations, emerging threats, and open research challenges. Our findings reveal significant vulnerabilities in LLM-based evaluation frameworks, as well as promising directions for improving their robustness and reliability. Finally, we outline key research opportunities that can guide the development of more secure and trustworthy LLM-as-a-Judge systems.

Aiman Al Masoud, Antony Anju, Marco Arazzi +6

Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy · Department of Computer Applications, Cochin University of Science and Technology, Kerala, India

LLM-as-a-JudgeLLM SecurityLLM ReliabilityAdversarial Attacks on LLMs
AskPDF
Mar 31, 2026

Audio Hallucination Attacks: Probing the Reliability of Large Audio Language Models

Large Audio Language Models (LALMs) achieve strong performance on audio-language tasks; however, their reliability in real-world settings remains underexplored. We introduce Audio Hallucination Attacks (AHA), an attack suite called AHA-Eval, comprising 6.5K QA pairs designed to test whether LALMs genuinely ground their responses in the audio input. AHA targets two attack surfaces: (i) query-based attacks, which exploit question structure to induce hallucinations about absent sounds, and (ii) audio-based attacks, which inject synthetic speech describing non-existent events into the audio stream. Evaluating state-of-the-art LALMs, including Audio Flamingo 3 and Gemini 3 Pro, we observe high attack success rates of 95.35% and 79.65%, respectively, revealing a reliability gap that is hidden by standard benchmark performance. To mitigate this, we propose a 120K QA post-alignment dataset, AHA-Guard, which successfully reduces attack success rates by up to 49%.

Ashish Seth, Sonal Kumar, Ramaneswaran Selvakumar +5

University of Maryland, College Park · Adobe Research

Audio-Language Model EvaluationLLM ReliabilityAudio-Language Model Hallucination Detection
AskPDF
Mar 31, 2026

Beyond pass@1: A Reliability Science Framework for Long-Horizon LLM Agents

Existing benchmarks measure capability -- whether a model succeeds on a single attempt -- but production deployments require reliability -- consistent success across repeated attempts on tasks of varying duration. We show these properties diverge systematically as task duration grows, and that pass@1 on short tasks is structurally blind to this divergence. We introduce a reliability science framework for long-horizon LLM agents with four metrics: Reliability Decay Curve (RDC), Variance Amplification Factor (VAF), Graceful Degradation Score (GDS), and Meltdown Onset Point (MOP). We evaluate 10 models across 23,392 episodes on a 396-task benchmark spanning four duration buckets and three domains. Key findings: (1) reliability decay is domain-stratified -- SE GDS drops from 0.90 to 0.44 while document processing is nearly flat (0.74 to 0.71); (2) VAF bifurcates by capability tier -- high VAF is a capability signature, not an instability signal; (3) capability and reliability rankings diverge substantially, with multi-rank inversions at long horizons; (4) frontier models have the highest meltdown rates (up to 19%) because they attempt ambitious multi-step strategies that sometimes spiral; and (5) memory scaffolds universally hurt long-horizon performance across all 10 models. These results motivate reliability as a first-class evaluation dimension alongside capability.

Aaditya Khanal, Yangyang Tao, Junxiu Zhou

School of Computing and Analytics, Northern Kentucky University

LLM ReliabilityLLM Agent ReliabilityLong-Horizon Agent EvaluationLong-Horizon LLM Agents
AskPDF
Mar 30, 2026

The Model Says Walk: Measuring whether LLMs Condition on Hidden Constraints

Asked whether to walk or drive to a car wash 50 m away, most language models say walk, forgetting that the car has to be there. Such failures are usually measured by accuracy on questions where the hidden constraint applies. We show that this is misleading: a model can answer these questions correctly without reasoning about the constraint at all, simply by favoring the cautious option. We instead ask whether a model's decision changes when the constraint is removed. We test this at three levels of control: log-probability sweeps on open-weight models, a 500-prompt stress test, and CORE, a new human-validated benchmark of minimal constraint-present/constraint-absent pairs. The picture is consistent. The constraint nudges decisions rather than governing them. Models still miss presence constraints like the car wash, yet elsewhere they apply constraints that are not there. As a result, standard accuracy flatters all ten models we evaluate and reorders their ranking, and prompting fixes that look effective largely vanish under paired scoring. Claims about hidden-constraint reasoning need paired evidence.

Yubo Li, Lu Zhang, Tianchong Jiang +2

Carnegie Mellon University · Independent Researcher

Constraint-Guided Reasoning in Language ModelsImplicit Reasoning in Language ModelsLLM EvaluationLLM Reliability
AskPDF