Daily Papers

In Agentic AI, Large Language Models (LLMs) are increasingly used in the orchestration layer to coordinate multiple agents and to interact with external services, retrieval components, and shared memory. In this setting, failures are not limited to incorrect final outputs. They also arise from long-horizon interaction, stochastic decisions, and external side effects (such as API calls, database writes, and message sends). Common failures include non-termination, role drift, propagation of unsupported claims, and attacks via untrusted context or external channels. This paper presents an assurance framework for such Agentic AI systems. Executions are instrumented as Message-Action Traces (MAT) with explicit step and trace contracts. Contracts provide machine-checkable verdicts, localize the first violating step, and support deterministic replay. The framework includes stress testing, formulated as a budgeted counterexample search over bounded perturbations. It also supports structured fault injection at service, retrieval, and memory boundaries to assess containment under realistic operational faults and degraded conditions. Finally, governance is treated as a runtime component, enforcing per-agent capability limits and action mediation (allow, rewrite, block) at the language-to-action boundary. To support comparative evaluations across stochastic seeds, models, and orchestration configurations, the paper defines trace-based metrics for task success, termination reliability, contract compliance, factuality indicators, containment rate, and governance outcome distributions. More broadly, the framework is intended as a common abstraction to support testing and evaluation of multi-agent LLM systems, and to facilitate reproducible comparison across orchestration designs and configurations.

GPU training jobs fail often, roughly two in five on large production clusters, yet the operator typically learns of a failure only by reconnecting hours later. Experiment trackers require editing the training script and maintaining a cloud connection; the scheduler's mail hook delivers a single status line with no cause and no logs. GPUAlert is a command-line wrapper that monitors any training command at the process boundary, and with no change to that command, emails a structured notification on completion carrying a classified failure cause, durable logs, and output artifacts. The tool is organized around three reliability primitives: a pre-launch log guarantee that establishes the durable destination before the child process can crash, notifier isolation that makes the wrapper's exit code a pure function of the child's status regardless of whether the email succeeds, and a non-silent artifact budget that bounds attachment size without ever dropping output silently. We release a labelled corpus of 474 GPU training logs across 15 failure classes and a reproducible evaluation harness. On the twelve hardware-reproduced classes, the ordered-rule classifier reaches 0.997 macro-F1, against 0.830 for unordered keyword matching and 0.133 for exit-code inspection. Wrapper overhead is a constant approximately 3ms per job; the pre-launch guarantee preserves a log where a shell redirect yields nothing; and across all 15 failure modes the wrapper returns the child's exit code unchanged even when the SMTP relay is unreachable.

Serial verification gates are a core reliability primitive in LLM harnesses: a candidate answer is returned only if k verifier calls all accept it. Under conditionally independent gates, the recent Odds Law (arXiv:2606.15712) shows that posterior log-odds grow linearly in k, so failure decays exponentially, and states that "a tight theory of partially correlated verifier cascades remains open." This note gives a minimal such theory. Modeling the per-instance false-accept rate on the generator's own errors as a latent variable αsim G (de Finetti), the exact cascade posterior is ell_k = ell_0 - ln m_k, with m_k the k-th moment of G. Then: (i) ell_k is concave in k for every non-degenerate G -- the Odds Law is its tangent at the first gate and an upper bound; (ii) for Beta(a,b) latents, failure decays polynomially, 1-r_k asymp k^{-b}, with correlation parameter ρ_v = 1/(a+b+1); (iii) a blind-spot atom of mass 1-π at α=1 caps the evidence extractable from any number of gates at -ln(1-π) nats, so reliability saturates below 1; (iv) letting the true-accept rate also vary (βsim H) yields a trichotomy -- gates eventually always help, plateau, or actively harm -- decided by the upper-tail exponents of G and H, with closed-form crossover k^dagger. The mechanism is survivorship: errors surviving gates are the high-α ones. The theory is measurable: R repeated verdicts per instance identify the first R moments of G, so two verdicts identify ρ_v; beta-binomial likelihood and NPMLE recover the reliability curve and the ill-posed ceiling. In synthetic tests, independence-based extrapolation underestimates failure by 20x at k=5 and ~3000x at k=10; the correlated fit at R=8 tracks held-out depths. The practical lever is decorrelation -- changing model family, modality, or evidence source -- not adding gates.

Given a black-box AI system and a task, at what confidence level can a practitioner trust the system's output? We answer with a reliability level -- a single number per system-task pair, derived from self-consistency sampling and conformal calibration, that serves as a black-box deployment gate with exact, finite-sample, distribution-free guarantees. Self-consistency sampling reduces uncertainty exponentially; conformal calibration guarantees correctness within 1/(n+1) of the target level, regardless of the system's errors -- made transparently visible through larger answer sets for harder questions. Weaker models earn lower reliability levels (not accuracy -- see Definition 2.4): GPT-4.1 earns 94.6% on GSM8K and 96.8% on TruthfulQA, while GPT-4.1-nano earns 89.8% on GSM8K and 66.5% on MMLU. We validate across five benchmarks, five models from three families, and both synthetic and real data. Conditional coverage on solvable items exceeds 0.93 across all configurations; sequential stopping reduces API costs by around 50%.

Existing benchmarks for tool-using LLM agents primarily report single-run success rates and miss reliability properties required in production. We introduce ReliabilityBench, a benchmark for evaluating agent reliability across three dimensions: (i) consistency under repeated execution using pass^k, (ii) robustness to semantically equivalent task perturbations at intensity ε, and (iii) fault tolerance under controlled tool/API failures at intensity λ. ReliabilityBench contributes a unified reliability surface R(k,ε,λ), action metamorphic relations that define correctness via end-state equivalence rather than text similarity, and a chaos-engineering-style fault injection framework (timeouts, rate limits, partial responses, schema drift). We evaluate two models (Gemini 2.0 Flash, GPT-4o) and two agent architectures (ReAct, Reflexion) across four domains (scheduling, travel, customer support, e-commerce) over 1,280 episodes. Perturbations alone reduce success from 96.9% at ε=0 to 88.1% at ε=0.2. Rate limiting is the most damaging fault in ablations. ReAct is more robust than Reflexion under combined stress, and Gemini 2.0 Flash achieves comparable reliability to GPT-4o at much lower cost. ReliabilityBench provides a systematic framework for assessing production readiness of 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.

Multi-step enterprise agent tasks fail in a characteristic way: single-pass inference has no checkpoint between deciding an answer and committing to it. We study one production system (Leni) whose architecture installs such checkpoints: verification loops (execute, observe, compare, correct) staffed by lightweight task-specialized post-trained models. We evaluate the unmodified production configuration on three public benchmarks stressing distinct failure modes: SpreadsheetBench Verified (silent computation error), BullshitBench v2 (premise confabulation), and the GAIA validation split (cascade error over long tool chains). The full system improves over its frontier base model by +11.0 percentage points on SpreadsheetBench (91.25% vs 80.25%, n=400, p<0.001), +7 to +10 percentage points on BullshitBench (98% vs 91%, n=100), and roughly +15 points on GAIA validation (75.2% pass@1, n=165; 83.0% best-of-k). Our central contribution is a decomposition of that uplift: most of it comes from scaffolding, routing, and specialist models rather than from the verification step itself, whose isolated contribution is small (+1.5 points) but concentrated at the top of the score distribution, where it converts otherwise-failing tasks. We instrument the loop end-to-end, yielding an empirical verifier confusion matrix (catch rate about 0.20, fix rate 0.75, no false-alarm regressions) that grounds a compounding-reliability model. Specialist-swap ablations suggest that the loop's value depends on who observes it: replacing the small trained verifier with the generating frontier model eliminates most rescues. A valid-premise control shows zero over-rejections in 100 expert-level questions.

Reasoning failures in large language models (LLMs) are typically measured only at the end of a generation, yet many failures manifest as a process-level breakdown: the model "loses the thread" mid-reasoning. We study whether such breakdowns are detectable from inference-time observables available in standard APIs (token log probabilities), without any training or fine-tuning. We define a simple instability signal that combines consecutive-step distributional shift (JSD) and uncertainty (entropy), summarize each trace by its peak instability strength, and show that this signal reliably predicts failure. Across GSM8K and HotpotQA, instability strength predicts wrong answers with above-chance AUC and yields monotonic bucket-level accuracy decline at scale across model sizes. Crucially, we show that instability is not uniformly harmful: early instability can reflect subsequent stabilization and a correct final answer (corrective instability), whereas late instability is more often followed by failure (destructive instability), even at comparable peak magnitudes, indicating that recoverability depends not only on how strongly the distribution changes but also on when such changes occur relative to the remaining decoding horizon. The method is model-agnostic, training-free, and reproducible, and is presented as a diagnostic lens rather than a corrective or control mechanism.

Counterexample-guided repair aims at creating neural networks with mathematical safety guarantees, facilitating the application of neural networks in safety-critical domains. However, whether counterexample-guided repair is guaranteed to terminate remains an open question. We approach this question by showing that counterexample-guided repair can be viewed as a robust optimisation algorithm. While termination guarantees for neural network repair itself remain beyond our reach, we prove termination for more restrained machine learning models and disprove termination in a general setting. We empirically study the practical implications of our theoretical results, demonstrating the suitability of common verifiers and falsifiers for repair despite a disadvantageous theoretical result. Additionally, we use our theoretical insights to devise a novel algorithm for repairing linear regression models based on quadratic programming, surpassing existing approaches.

Corruption studies, the primary tool for evaluating chain-of-thought (CoT) faithfulness, identify which chain positions are "computationally important" by measuring accuracy when steps are replaced with errors. We identify a systematic confound: for chains with explicit terminal answer statements, the dominant format in standard benchmarks, corruption studies detect where the answer text appears, not where computation occurs. A within-dataset format ablation provides the key evidence: on standard GSM8K chains ending with "the answer is X," removing only the answer statement, preserving all reasoning, collapses suffix sensitivity ~19x at 3B (N=300, p=0.022). Conflicting-answer experiments quantify the causal mechanism: at 7B, CC accuracy drops to near-zero (<=0.02) across five architecture families; the followed-wrong rate spans 0.63-1.00 at 3B-7B and attenuates at larger scales (0.300 at Phi-4-14B, ~0.01 at 32B). A within-stable 7B replication (9.3x attenuation, N=76, p=7.8e-3; Qwen3-8B N=299, p=0.004) provides converging evidence, and the pattern replicates on MATH (DeepSeek-R1-7B: 10.9x suffix-survival recovery). On chains without answer suffixes the same protocol identifies the prefix as load-bearing (Delta=-0.77, p<10^-12). Generation-time probes confirm a dissociation: the answer is not early-determined during generation (early commitment <5%), yet at consumption time model outputs systematically follow the explicit answer text. The format-determination effect persists through 14B (8.5x ratio, p=0.001) and converges toward zero at 32B. We propose a three-prerequisite protocol (question-only control, format characterization, all-position sweep) as a minimum standard for corruption-based faithfulness studies.

How do multi-turn reasoning systems fail? The expected answer is logical contradiction, in which the system's maintained state becomes unsatisfiable. We show that the dominant mode is instead satisfiable drift, where the internal state stays consistent while the returned answer silently violates prior commitments. We build DRIFT-Bench (Decomposing Reasoning Into Failure Types), a solver-instrumented benchmark of 816 test problems across three constraint domains, and evaluate four methods on it across four open-weight models (8B-120B parameters). MUS-Repair, which feeds minimal unsatisfiable subsets back to the generator, is strongest in every setting (+1.8 to +15.0 pp over the best non-MUS baseline). But the central finding is what repair leaves behind. After structured feedback, models rarely contradict themselves. They forget. Residual errors are 98-100% satisfiable drift across all settings, while contradiction drops to near zero. Reliable multi-turn systems must separately validate that the returned answer respects the maintained state. Code is available at https://github.com/kaons-research/drift-bench.

Self-evolving agent systems improve by creating, revising, and retiring their own skills, but every such loop rests on a hidden assumption: a reliable evaluation metric already exists. In many real applications it does not. We make three claims. First, metrics can be evolved: our metric loop searches compositions of small drawback detectors under a full evolutionary lifecycle, trained to agree with a ten-item anchored reference set, regularized by consensus over unlabeled outputs, and audited against a held-out anchor it never reads, yielding a transparent, inspectable metric rather than an opaque judge. Second, since no metric exists to beat, the yardstick is recovering what an accurate metric would have enabled, and Double Ratchet, our co-evolution of the metric with a lifecycle-managed skill loop, does so: across code generation (MBPP+), enterprise text-to-SQL (Spider~2.0-Snow), and reference-free report generation, it retains 88--110\% of the held-out lift achieved by the same skill loop driven by ground truth or the best available rubric. Third, safety comes from anchor discipline plus outer audits: removing anchor guards collapses the metric into a vacuous detector while removing the lifecycle does not; and when evolved skills gamed the report rubric, an independent judge caught it, one detector repaired it, and a task-aware judge then preferred the evolved outputs over the pre-evolution baseline in 77\% of decided pairs. We argue this failure-expecting architecture is the right default wherever no reliable automatic verifier exists.

Agentic LLM coding tools compress long session histories into compaction summaries that subsequent sessions inherit as ground truth. This paper documents a failure mode in Claude Code where partial standard output from timed-out commands (exit code 143) is recorded in compaction summaries as confirmed results, propagating false positives across sessions and model versions without re-verification. The underlying mechanism is a conflation of observation and persistence, where information that appeared in the terminal is treated as equivalent to information written to durable storage. This finding extends the analysis of LLM self-evaluation failures reported in prior work on non-determinism in LLM-as-judge grading by showing that agentic tools exhibit analogous reliability deficits when reporting on their own operational outcomes. The failure has direct implications for any workflow that relies on agentic session continuity for data processing, scientific computation, or multi-step automation.

Evaluating the reliability of news sources is a routine task for journalists and organizations committed to acquiring and disseminating accurate information. Recent research has shown that predicting sources' reliability represents an important first-prior step in addressing additional challenges such as fake news detection and fact-checking. In this paper, we introduce a novel approach for source reliability estimation that leverages reinforcement learning strategies for estimating the reliability degree of news sources. Contrary to previous research, our proposed approach models the problem as the estimation of a reliability degree, and not a reliability label, based on how all the news media sources interact with each other on the Web. We validated the effectiveness of our method on a news media reliability dataset that is an order of magnitude larger than comparable existing datasets. Results show that the estimated reliability degrees strongly correlates with journalists-provided scores (Spearman=0.80) and can effectively predict reliability labels (macro-avg. F_1 score=81.05). We release our implementation and dataset, aiming to provide a valuable resource for the NLP community working on information verification.

Handwritten STEM exams capture open-ended reasoning and diagrams, but manual grading is slow and difficult to scale. We present an end-to-end workflow for grading scanned handwritten engineering quizzes with multimodal large language models (LLMs) that preserves the standard exam process (A4 paper, unconstrained student handwriting). The lecturer provides only a handwritten reference solution (100%) and a short set of grading rules; the reference is converted into a text-only summary that conditions grading without exposing the reference scan. Reliability is achieved through a multi-stage design with a format/presence check to prevent grading blank answers, an ensemble of independent graders, supervisor aggregation, and rigid templates with deterministic validation to produce auditable, machine-parseable reports. We evaluate the frozen pipeline in a clean-room protocol on a held-out real course quiz in Slovenian, including hand-drawn circuit schematics. With state-of-the-art backends (GPT-5.2 and Gemini-3 Pro), the full pipeline achieves approx8-point mean absolute difference to lecturer grades with low bias and an estimated manual-review trigger rate of approx17% at D_{max}=40. Ablations show that trivial prompting and removing the reference solution substantially degrade accuracy and introduce systematic over-grading, confirming that structured prompting and reference grounding are essential.

Reliable application of machine learning is of primary importance to the practical deployment of deep learning methods. A fundamental challenge is that models are often unreliable due to overconfidence. In this paper, we estimate a model's reliability by measuring the agreement between its latent space, and the latent space of a foundation model. However, it is challenging to measure the agreement between two different latent spaces due to their incoherence, \eg, arbitrary rotations and different dimensionality. To overcome this incoherence issue, we design a neighborhood agreement measure between latent spaces and find that this agreement is surprisingly well-correlated with the reliability of a model's predictions. Further, we show that fusing neighborhood agreement into a model's predictive confidence in a post-hoc way significantly improves its reliability. Theoretical analysis and extensive experiments on failure detection across various datasets verify the effectiveness of our method on both in-distribution and out-of-distribution settings.

Deployed language models must produce outputs that are both correct and format-compliant. We study this structured-output reliability gap using two mathematical benchmarks -- GSM8K and MATH -- as a controlled testbed: ground truth is unambiguous and the output contract is strict (JSON with required fields). We evaluate three 7-9B models under five prompting strategies and report output accuracy -- the joint event of mathematical correctness and valid JSON structure -- as the primary metric. A systematic format failure emerges: NAIVE prompting (no system prompt) achieves up to 85% task accuracy on GSM8K but 0% output accuracy across all models and datasets. REFERENCE prompting (a minimal hand-written JSON format prompt) fares little better, yielding 0% output accuracy for two of four models tested. Constrained decoding enforces syntactic validity but incurs 3.6x-8.2x latency overhead and in several settings degrades task performance substantially. To overcome this limitation, we developed AloLab, an iterative system-prompt optimizer (meta-agent: Claude Sonnet 4.5) requiring only black-box API access to the target model; it reaches 84-87% output accuracy on GSM8K and 34-40% on MATH across five independent runs per model, with 29/30 paired McNemar comparisons against the best static prompt significant at p < 0.05, at near-NAIVE inference latency and without model fine-tuning. The same format failure extends to GPT-4o (OpenAI, 2024), a proprietary closed-source model: REFERENCE achieves 0% output accuracy due to systematic markdown-fence wrapping, while AloLab reaches 95.2% [94.8, 95.6]. An ablation replacing the Sonnet 4.5 meta-agent with Claude 3 Haiku reduces mean output accuracy to 61.0% and increases run-to-run standard deviation from <1 pp to 21.8 pp, confirming that meta-agent capability is a primary driver of optimization quality.

Advanced LLMs have achieved near-ceiling instruction-following accuracy on benchmarks such as IFEval. However, these impressive scores do not necessarily translate to reliable services in real-world use, where users often vary their phrasing, contextual framing, and task formulations. In this paper, we study nuance-oriented reliability: whether models exhibit consistent competence across cousin prompts that convey analogous user intents but with subtle nuances. To quantify this, we introduce a new metric, reliable@k, and develop an automated pipeline that generates high-quality cousin prompts via data augmentation. Building upon this, we construct IFEval++ for systematic evaluation. Across 20 proprietary and 26 open-source LLMs, we find that current models exhibit substantial insufficiency in nuance-oriented reliability -- their performance can drop by up to 61.8% with nuanced prompt modifications. What's more, we characterize it and explore three potential improvement recipes. Our findings highlight nuance-oriented reliability as a crucial yet underexplored next step toward more dependable and trustworthy LLM behavior. Our code and benchmark are accessible: https://github.com/jianshuod/IFEval-pp.

State machine replication (SMR) and Byzantine fault-tolerant (BFT) consensus guarantee agreement despite a bounded number of arbitrary, colluding faulty participants. However, these guarantees rely on participants outside this set correctly executing the protocol's transition semantics. Agentic validators expose a weaker boundary: an authenticated, responsive, non-equivocating, and protocol-compliant reasoning participant may still endorse a semantically invalid transition due to reasoning errors. We call this failure mode an epistemic fault, and the collective phenomenon the Honest Quorum Problem (where "honest" means protocol-compliant, not semantically correct). Such a quorum can satisfy ordinary checks while forming a certificate for an invalid transition. Thus, agreement alone does not guarantee semantic validity or execution safety. Furthermore, because agentic validators often share model weights, training distributions, prompts, or toolchains, they are highly susceptible to correlated epistemic faults. We define Epistemic Byzantine Fault Tolerance (EBFT), a fault-tolerance model for agentic infrastructure and post-deterministic distributed systems. EBFT augments the conventional Byzantine fault bound with two separate, confidence-indexed quantities: e_δ bounds coherent invalid endorsements outside the Byzantine set, and u_ε bounds unusable validator support that degrades liveness. These quantities characterize semantic safety risk and liveness degradation independently. We derive quorum-threshold conditions for semantic validity, consensus agreement, liveness, and feasible threshold selection, and outline a calibration methodology for estimating these budgets. We show that adding nominally distinct agents improves fault tolerance only when it measurably reduces the upper-tail concentration of invalid endorsements or unusable support.

Large language models are increasingly deployed as agents that solve tasks by interacting with external tool environments. Although recent tool-use benchmarks increasingly cover complex task settings, they still largely assume clean, stable, and trustworthy tool environments, leaving tool-environment unreliability insufficiently examined. We introduce ToolBench-X, a benchmark for evaluating agents under recoverable reliability hazards. ToolBench-X contains executable multi-step tasks across diverse domains and sequential, parallel, and mixed workflows, each paired with deterministic tools and a canonical final answer for automatic evaluation. Starting from clean tool environments, ToolBench-X injects five structured hazard types: Specification Drift, Invocation Error, Execution Failure, Output Drift, and Cross-source Conflict. Crucially, each injected instance remains solvable through at least one valid recovery path, such as retrying, fallback, verification, or cross-checking. Experiments reveal a substantial reliability gap: agents that perform well with reliable tools often fail under recoverable hazards. Further analysis shows that failures are driven less by tool-use volume or inference budget than by limited hazard diagnosis and ineffective recovery. Targeted recovery hints recover many failed tasks, while test-time scaling yields more limited gains. These results suggest that tool-use evaluation should move beyond function-call accuracy toward task completion under unreliable tool environments. The code and data is available at https://github.com/Foreverskyou/ToolBench-X.

Red-team evaluations of AI models support some claims and not others, and the boundary between the two is calculable rather than merely a matter of judgment. We define the evidential ceiling of an evaluation as the largest factor by which one result can move belief under a fixed testing budget, derive it in closed form for the benchmark null result, and use it to locate that boundary exactly. We find that above a calculable harm rate, a benchmark of modest size certifies a category to a stated evidentiary standard, and a clean sheet is then the stronger of the two possible observations, outweighing a single reproduced failure. Below that rate, no passive benchmark of feasible size provides the specified evidence of safety under the fixed scoring rule and approximately independent trial structure. The crossing between the two regimes has a closed form. The bound is not specific to benchmarks: written in terms of a procedure's hypothesis conditioned elicitation rates, it covers adaptive and automated red teaming as well, and shows that discrimination between the hypotheses rather than attack success is what determines evidential worth. Auditing eight evaluation suites against the boundary, we find that current benchmarks are adequate for high-frequency harm categories and several orders of magnitude short for rare, catastrophic ones. Safety benchmarks are not uninformative. They are informative about a specific and computable set of propositions, and the discipline they need is to state which.

Single-prompt accuracy is the dominant way to benchmark language models, but it can miss reliability failures that matter. We evaluate a 15-model open-weight corpus, with the main reliability analyses focused on 10 instruct models across five classification and reasoning benchmarks under five prompt variants each, measuring accuracy, token-probability calibration, verbal-confidence calibration, verbal parse rate, and prompt-perturbation spread for every (model x dataset x variant) cell. We find three broad results. First, evaluation design can materially change the conclusion. Switching Expected Calibration Error (ECE) token from a raw to a label-set-normalised definition changes per-cell calibration by a mean absolute 0.149. More strikingly, pairing a chain-of-thought prompt with a first-character evaluator on ARC-Challenge reduces apparent accuracy by 72-88% across all five primary models; two independent repair procedures recover 93.8% and 102.7% of the lost performance, indicating an evaluator-side rather than model-side failure. Second, confidence signals are fragile. On MMLU-Pro, every primary model verbally reports confidence substantially above both its accuracy and its token-probability confidence on the same rows, and verbal parse rate can collapse for a single model on a single prompt variant. Third, prompt robustness does not track parameter count reliably. Across 10 instruct models, the correlation between model size and prompt-perturbation spread ranges from -0.244 to 0.474 across benchmarks. Taken together, these results show that reliability conclusions for small language models depend not only on the model being evaluated, but also on the evaluation pipeline used to measure it. We argue that calibration definitions, evaluator logic, verbal parseability, and prompt robustness should be reported explicitly when making reliability claims.

Large language models can arrive at the same answer through reasoning paths that are unstable, contradictory, or difficult to rank consistently -- a failure mode especially prevalent in multi-step deductive reasoning. Existing methods assess reliability primarily through output dispersion -- measuring how much sampled answers differ -- but this discards a complementary signal: whether the model can consistently rank competing reasoning candidates. We propose structural uncertainty, a consistency-aware framework derived from the stability of self-preference-induced rankings over sampled reasoning solutions. Given a query, we generate multiple candidate solutions and ask the model to judge pairwise preferences among its own outputs. We aggregate self-preferences into ranking distributions via Bradley-Terry modeling with PageRank, and decompose the signal into two entropy-based components: across-trial ranking instability and within-trial candidate ambiguity. Across five LLMs and eight benchmarks, structural signals provide information complementary to answer dispersion: on logical and mathematical reasoning tasks, the combination improves identification of unreliable instances, while on factual retrieval the structural signal collapses toward uniformity, diagnosing a regime boundary where reasoning-level consistency evaluation is uninformative. The two components relate differently to accuracy: within-trial ambiguity correlates positively with correctness -- consistent with settings where multiple plausible solution paths remain competitive -- while across-trial instability correlates negatively, signaling unreliable reasoning. Structural uncertainty is best understood not as a universal confidence estimator, but as a regime-sensitive evaluator of logical reasoning consistency.

Large Language Models often improve accuracy on reasoning tasks by sampling multiple Chain-of-Thought (CoT) traces and aggregating them with majority voting (MV), a test-time technique called self-consistency. When we truncate a CoT partway through and regenerate the remainder, we observe that traces with correct answers reproduce their original answer more often than traces with wrong answers. We use this difference as a reliability signal, prefix consistency, that weights each candidate answer by how often it reappears under regeneration. It requires no access to token log-probabilities or self-rating prompts. Across five reasoning models and four math and science benchmarks, prefix consistency is the best correctness predictor in most settings, and reweighting votes by it reaches Standard MV plateau accuracy at up to 21x fewer tokens (median 4.6x). Our code is available at https://github.com/naoto-iwase/prefix-consistency.

As LLM-based agents are deployed in production systems, understanding their behavioral consistency (whether they produce similar action sequences when given identical tasks) becomes critical for reliability. We study consistency in the context of SWE-bench, a challenging software engineering benchmark requiring complex, multi-step reasoning. Comparing Claude~4.5~Sonnet, GPT-5, and Llama-3.1-70B across 50 runs each (10 tasks times 5 runs), we find that across models, higher consistency aligns with higher accuracy: Claude achieves the lowest variance (CV: 15.2\%) and highest accuracy (58\%), GPT-5 is intermediate (CV: 32.2\%, accuracy: 32\%), and Llama shows the highest variance (CV: 47.0\%) with lowest accuracy (4\%). However, within a model, consistency can amplify both correct and incorrect interpretations. Our analysis reveals a critical nuance: consistency amplifies outcomes rather than guaranteeing correctness. 71\% of Claude's failures stem from "consistent wrong interpretation": making the same incorrect assumption across all runs. Interestingly, GPT-5 achieves similar early strategic agreement as Claude (diverging at step 3.4 vs.\ 3.2) but exhibits 2.1times higher variance, suggesting that divergence timing alone does not determine consistency. These findings suggest that for production deployment, interpretation accuracy matters more than execution consistency, with implications for agent evaluation and training.

A self-evolving agent retires its bad skills by watching them fail, so what happens when the judge cannot see the failures? Skill retirement is the structural constraint that keeps a growing library from drifting below the no-skill baseline, but its guarantee assumes an unbiased reward, which is false for the LLM judges that reference-free tasks force upon us. We show that a biased judge does not merely add noise; it silently switches off the curator. We make this precise with a corrupted-reward analysis and, isolating the causal channel by injecting corruption on top of a deterministic reward, a behavioral study on a reference-free report-writing testbed with a code-generation cross-check. Symmetric noise leaves retirement intact, but false-pass bias (failures slipping through as passes) disables contribution-based retirement past a sharp threshold that no amount of data can cross. Separating genuine retirement from cap-eviction churn shows this mechanism failure is universal, holding across domains and failure rates and sparing only near-zero-false-pass, verifier-like graders. The downstream outcome, though, is regime-dependent: eval quality degrades only where the same corruption also starves skill synthesis, and otherwise holds steady, so the disabled curator is silent, surfacing in no aggregate metric. The contribution is a behavioral safety result, not a performance one. A cheap defect-injection audit then tells an operator, before deployment, which side of the threshold their judge occupies.

Retrieval-Augmented Generation (RAG) is an effective approach to enhance the factual accuracy of large language models (LLMs) by retrieving information from external databases, which are typically composed of diverse sources, to supplement the limited internal knowledge of LLMs. However, the standard RAG often risks retrieving incorrect information, as it relies solely on relevance between a query and a document, overlooking the heterogeneous reliability of these sources. To address this issue, we propose Reliability-Aware RAG (RA-RAG), a new multi-source RAG framework that estimates the reliability of sources and leverages this information to prioritize highly reliable and relevant documents, ensuring more robust and accurate response generation. Specifically, RA-RAG first estimates source reliability by cross-checking information across multiple sources. It then retrieves documents from the top-kappa reliable and relevant sources and aggregates their information using weighted majority voting (WMV), where the selective retrieval ensures scalability while not compromising the performance. Comprehensive experiments show that RA-RAG consistently outperforms baselines in scenarios with heterogeneous source reliability while scaling efficiently as the number of sources increases. Furthermore, we demonstrate the ability of RA-RAG to estimate real-world sources' reliability, highlighting its practical applicability. Our code and data are available at \href{https://github.com/ml-postech/RA-RAG{RA-RAG}.}

Long-lived AI agents are increasingly deployed as persistent operational systems, yet they are still evaluated like freshly initialized models. Day-one benchmarks miss a basic systems question: how long does an agent remain reliable after deployment? Even when model weights are frozen, an agent's effective state keeps changing as it compresses interaction history, retrieves from a growing memory store, revises facts after updates, and undergoes routine maintenance. Reliability therefore becomes a lifespan property of the full agent harness, not only a snapshot property of the base model. We introduce AgingBench, a longitudinal reliability benchmark for agent lifespan engineering: measuring not only whether deployed agents degrade, but what form the degradation takes and where repair should target. AgingBench organizes agent aging into four mechanisms: compression aging, interference aging, revision aging, and maintenance aging. To diagnose these failures, AgingBench uses temporal dependency graphs and paired counterfactual probes that produce diagnostic profiles for the write, retrieval, and utilization stages of the memory pipeline. Across 7 scenarios, 14 models, multiple memory policies, and both runner-controlled and autonomous agents, over ~400 runs spanning 8 - 200 sessions show that agent aging is not one-dimensional: behavioral tests can remain clean while factual precision decays; derived-state tracking can collapse sharply within a single model; and the same wrong answer can require different repairs depending on what the diagnostic profile points to. These results suggest that reliable agent deployment requires lifespan evaluation, mechanism-level diagnosis, and stage-targeted repair, not only stronger day-one models.

Reliability is essential for ensuring efficiency in LLM training. However, many real-world reliability issues remain difficult to resolve, resulting in wasted resources and degraded model performance. Unfortunately, today's collective communication libraries operate as black boxes, hiding critical information needed for effective root cause analysis. We propose Mycroft, a lightweight distributed tracing and root cause analysis system designed to address previously hidden reliability issues in collective communication. Mycroft's key idea is to trace collective communication states and leverage internal control and data dependencies to resolve reliability problems in LLM training. Mycroft has been deployed at ByteDance for over six months to debug collective communication related issues at runtime. It detected anomalies within 15 seconds in 90% of cases and identified the root cause within 20 seconds in 60% of cases. We also conducted extensive fault injection experiments to demonstrate Mycroft's capability and efficiency.

Large Reasoning Models (LRMs) exhibit strong performance, yet often produce rationales that sound plausible but fail to reflect their true decision process, undermining reliability and trust. We introduce a formal framework for reasoning faithfulness, defined by two testable conditions: stance consistency (a coherent stance linking reasoning to answer) and causal influence (the stated reasoning causally drives the answer under output-level interventions), explicitly decoupled from accuracy. To operationalize this, we present RFEval, a benchmark of 7,186 instances across seven tasks that probes faithfulness via controlled, output-level counterfactual interventions. Evaluating twelve open-source LRMs, we find unfaithfulness in 49.7% of outputs, predominantly from stance inconsistency. Failures are concentrated in brittle, convergent domains such as math and code, and correlate more with post-training regimes than with scale: within-family ablations indicate that adding current RL-style objectives on top of supervised fine-tuning can reduce reasoning faithfulness, even when accuracy is maintained. Crucially, accuracy is neither a sufficient nor a reliable proxy for faithfulness: once controlling for model and task, the accuracy-faithfulness link is weak and statistically insignificant. Our work establishes a rigorous methodology for auditing LRM reliability and shows that trustworthy AI requires optimizing not only for correct outcomes but also for the structural integrity of the reasoning process. Our code and dataset can be found at project page: https://aidaslab.github.io/RFEval/}{https://aidaslab.github.io/RFEval/

We study reliability in autonomous language-model agents that translate user mandates into validated tool actions under real capital. The setting is DX Terminal Pro, a 21-day deployment in which 3,505 user-funded agents traded real ETH in a bounded onchain market. Users configured vaults through structured controls and natural-language strategies, but only agents could choose normal buy/sell trades. The system produced 7.5M agent invocations, roughly 300K onchain actions, about $20M in volume, more than 5,000 ETH deployed, roughly 70B inference tokens, and 99.9% settlement success for policy-valid submitted transactions. Long-running agents accumulated thousands of sequential decisions, including 6,000+ prompt-state-action cycles for continuously active agents, yielding a large-scale trace from user mandate to rendered prompt, reasoning, validation, portfolio state, and settlement. Reliability did not come from the base model alone; it emerged from the operating layer around the model: prompt compilation, typed controls, policy validation, execution guards, memory design, and trace-level observability. Pre-launch testing exposed failures that text-only benchmarks rarely measure, including fabricated trading rules, fee paralysis, numeric anchoring, cadence trading, and misread tokenomics. Targeted harness changes reduced fabricated sell rules from 57% to 3%, reduced fee-led observations from 32.5% to below 10%, and increased capital deployment from 42.9% to 78.0% in an affected test population. We show that capital-managing agents should be evaluated across the full path from user mandate to prompt, validated action, and settlement.

Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully. In policy-permissive environments, a tool may execute any well-formed call even when the corresponding state transition is forbidden by domain policy. The result is a silent wrong state (a booking cancelled, a passenger count changed, a claim acted on without verification) that neither the tool nor the agent's self-report exposes. We study this failure mode in the τ^2-bench airline domain. On a budget agent, 78% of observed failures are silent wrong-state failures with no tool error, and the aggregate failure rate is reproducible across disjoint seeds, not sampling noise. We then evaluate a lightweight intervention: deterministic, read-only pre-execution gates that inspect the proposed call and current state before allowing a write. A four-gate suite raises full-benchmark success from 29.6% to 42.0% on gpt-4o-mini (+12.4pp; paired task-level bootstrap P=0.0012), and the lift reproduces on a disjoint 15-seed set (+12.3pp; P=0.0008). The effect is concentrated where the gates fire: on the 26/50 firing tasks, success rises by +19.2pp, while movement on the 24 non-firing tasks does not exclude zero. Two negative controls (a self-enforcing retail domain and BFCL) bound the mechanism: gates help when tools are policy-permissive and add little where tools already self-enforce. As suggestive evidence, not a central claim, the same failure mode persists at the frontier: gpt-5.2 at default reasoning still attempts policy-violating writes, and the same suite improves success from 61.2% to 71.6% (+10.4pp; P=0.020; n=5, no replication). The contribution is a bounded evaluation and reliability result: deterministic gates do not guarantee task success, but they can deterministically prevent a known class of silent policy-violating writes at the action boundary.

Large language model (LLM) coding agents are increasingly deployed to autonomously perform software engineering tasks in terminal-based environments, making their reliability a growing concern. Existing empirical studies investigate why coding agents fail, yet they largely treat failure as a final outcome rather than a temporal process, providing limited insight into how failures emerge, evolve, and become unrecoverable. We present the first large-scale empirical study of CLI coding-agent failure trajectories, introducing a process-oriented framework that analyzes failure through its onset, evolution, and recovery across execution trajectories. We first collect 3,843 execution trajectories generated by seven frontier models across three coding-agent scaffolds (OpenHands, MiniSWE, and Terminus2) on Terminal-Bench, then carefully filter them to obtain 1,794 complete and valid trajectories for manual annotation (over 63,000 execution steps), from which we derive 14 findings spanning failure occurrence, root causes, recovery, and cross-system consistency. Our findings show that coding-agent failures are predominantly driven by epistemic errors, typically begin within the first few execution steps, and often remain hidden until recovery is no longer possible, suggesting that improving coding-agent reliability requires earlier validation and intervention rather than relying solely on final-outcome evaluation.

What happens to an LLM agent's tool choice when the reliable tool silently changes within an ongoing session? We borrow set-shifting from cognitive psychology to study how well agents adapt to hidden reliability shifts. Our benchmark mounts tool-skill libraries with redundancies, where many tools solve the same task but differ in hidden reliability. In our evaluation framework, a branched schedule shifts the reliable tool group at hidden boundaries and pairs every shift with a no-shift control. We find that agents, by default, settle on a small recurring routine within a few turns of each boundary, with call shares concentrating on a few discrete values after each reliability shift. We score the set-shifting accuracy for each agent trajectory: the joint probability of routing to the target tool group in every post-shift window. We test open-weight LLMs in an open-source agentic harness and find qualitatively distinct failure modes across the same set of routines. We also find that set framing, how the toolset presents the alternatives as competing or complementary, shifts the routing dynamics.

Recent advances in long chain-of-thought (CoT) reasoning have largely prioritized answer accuracy and token efficiency, while overlooking aspects critical to trustworthiness. We argue that usable reasoning systems must be trustworthy, characterized by three properties: interpretability, faithfulness, and reliability. To this end, we propose ReFIne, a new training framework that integrates supervised fine-tuning with GRPO to encourage models to: (i) improve interpretability by producing structured, tag-based traces with high-level planning that are easier for humans to follow; (ii) enhance faithfulness by explicitly disclosing the decisive information guiding each solution, with consistent cross-section references; and (iii) promote reliability by providing self-assessments of both the derivation's soundness and the confidence of the final answer. We apply ReFIne to the Qwen3 models at multiple scales (1.7B/4B/8B) and evaluate across mathematical benchmarks of varying difficulty. Our experimental results show that ReFIne models generate clearer and better-structured reasoning traces (interpretability +44.0%), more faithfully expose their underlying decision process (faithfulness +18.8%), and offer informative confidence estimates (reliability +42.4%). These findings highlight an overlooked but important direction: reasoning models should be optimized not only for accuracy, but also for broader dimensions of trustworthiness. Our code is available at: https://github.com/Trustworthy-ML-Lab/Training_Trustworthy_LRM_with_Refine

Although demonstrating remarkable performance on reasoning tasks, Large Language Models (LLMs) still tend to fabricate unreliable responses when confronted with problems that are unsolvable or beyond their capability, severely undermining the reliability. Prior studies of LLM reliability have primarily focused on knowledge tasks to identify unanswerable questions, while mathematical reasoning tasks have remained unexplored due to the dearth of unsolvable math problems. To systematically investigate LLM reliability in mathematical reasoning tasks, we formulate the reliability evaluation for both solvable and unsolvable problems. We then develop a ReliableMath dataset which incorporates open-source solvable problems and high-quality unsolvable problems synthesized by our proposed construction workflow with human evaluations. Experiments are conducted on various LLMs with several key findings uncovered. LLMs fail to directly identify unsolvable problems and always generate fabricated responses. When instructing LLMs to indicate unsolvability using a reliable prompt, the reliability of larger-sized LLMs remains on solvable problems, but notably improves on unsolvable problems yet still falls short of solvable problems. However, small LLMs rarely show any progress despite employing reliable prompts. Therefore, we further propose an alignment strategy to enhance small LLMs' reliability, which can significantly improve LLM reliability performances on both in-domain and out-of-domain tasks.

Benchmarks measure whether a model is correct. They do not measure whether a model is reliable. This distinction is largely academic for single-shot inference, but becomes critical for agentic AI systems, where a single rephrased prompt can trigger cascading failures in multi-step execution. Yet this form of instability is not captured by existing evaluations. Hallucinations live in variance: they arise when semantically equivalent prompts activate inconsistent internal pathways, producing divergent outputs. Consistent but incorrect outputs reflect bias or missing knowledge; confident guessing reflects calibration failure. Neither constitutes hallucination under this definition. When error is variance-dominated, reducing redundant pathways improves reliability without adding knowledge. We formalize this through Semantic Stability (SS), measured via Paraphrase Consistency (PC@k): generate k paraphrases, greedy decode each, compute mode agreement. SS is a diagnostic for variance-driven unreliability, not a method for improving correctness. We show that a dense Qwen3-0.6B agrees with itself only 23.8% of the time; at 32% sparsity, agreement jumps to 55.9%. A phase diagram reveals the sweet spot where variance reduction outpaces bias accumulation, and regimes where stability collapses onto wrong answers.

Ensuring the reliability and safety of automated decision-making is crucial. It is well-known that data distribution shifts in machine learning can produce unreliable outcomes. This paper proposes a new approach for measuring the reliability of predictions under distribution shifts. We analyze how the outputs of a trained neural network change using clustering to measure distances between outputs and class centroids. We propose this distance as a metric to evaluate the confidence of predictions under distribution shifts. We assign each prediction to a cluster with centroid representing the mean softmax output for all correct predictions of a given class. We then define a safety threshold for a class as the smallest distance from an incorrect prediction to the given class centroid. We evaluate the approach on the MNIST and CIFAR-10 datasets using a Convolutional Neural Network and a Vision Transformer, respectively. The results show that our approach is consistent across these data sets and network models, and indicate that the proposed metric can offer an efficient way of determining when automated predictions are acceptable and when they should be deferred to human operators given a distribution shift.

In the age of open and free information, a concerning trend of reliance on AI is emerging. However, existing AI tools struggle to evaluate the credibility of information and to justify their assessments. Hence, there is a growing need for systems that can help users evaluate the trustworthiness of online information. Although major search engines incorporate AI features, they often lack clear reliability indicators. We present TrueGL, a model that makes trustworthy search results more accessible. The model is a fine-tuned version of IBM's Granite-1B, trained on the custom dataset and integrated into a search engine with a reliability scoring system. We evaluate the system using prompt engineering and assigning each statement a continuous reliability score from 0.1 to 1, then instructing the model to return a textual explanation alongside the score. Each model's predicted scores are measured against real scores using standard evaluation metrics. TrueGL consistently outperforms other small-scale LLMs and rule-based approaches across all experiments on key evaluation metrics, including MAE, RMSE, and R2. The model's high accuracy, broad content coverage, and ease of use make trustworthy information more accessible and help reduce the spread of false or misleading content online. Our code is publicly available at https://github.com/AlgazinovAleksandr/TrueGL, and our model is publicly released at https://huggingface.co/JoydeepC/trueGL.

Clinical decision-support systems are expert systems whose recommendations clinicians act on directly, yet they are usually cleared on one aggregate accuracy number from a held-out test set. That number says nothing about input reliability under encoding shifts, subgroup gaps, threshold sensitivity, or operational feasibility. We present RISED, a pre-deployment evaluation framework operationalising five dimensions (Reliability, Inclusivity, Sensitivity, Equity, Deployability) through BCa bootstrap 95% confidence intervals, literature-grounded thresholds, and Holm-Bonferroni-corrected PASS / FAIL / INCONCLUSIVE verdicts; Equity is a proxy-dependence diagnostic rather than a gating test. Applied to seven cohorts spanning 35 years (n from 303 to 99,492), RISED surfaces failures invisible to AUROC: on Diabetes 130, Reliability passes by three orders of magnitude (PSS = 0.0004) while Inclusivity (AUC parity gap = 0.262) and Sensitivity (max threshold-flip rate 49.1%) fail decisively; both NHIS cohorts reproduce this. NHANES 2021-2023, with a complete feature profile, achieves INCONCLUSIVE verdicts; BRFSS 2024 produces the suite's most severe Sensitivity failure (max threshold-flip rate 64.2%) after instrument rotation removed hypertension and cholesterol. The pattern recurs on credit- and income-prediction cohorts, confirming domain-agnosticity; a multi-model check shows the failures are data-driven, not model-specific. RISED ships as an open-source Python package complementing TRIPOD+AI, FUTURE-AI, and Fairlearn with the structured numerical evidence those standards require but do not prescribe.

Multimodal Large Language Models (MLLMs) have significantly advanced document understanding, yet current Doc-VQA evaluations score only the final answer and leave the supporting evidence unchecked. This answer-only approach masks a critical failure mode: a model can land on the correct answer while grounding it in the wrong passage -- a critical risk in high-stakes domains like law, finance, and medicine, where every conclusion must be traceable to a specific source region. To address this, we introduce CiteVQA, a benchmark that requires models to return element-level bounding-box citations alongside each answer, evaluating both jointly. CiteVQA comprises 1,897 questions across 711 PDFs spanning seven domains and two languages, averaging 40.6 pages per document. To ensure fidelity and scalability, the ground-truth citations are generated by an automated pipeline-which identifies crucial evidence via masking ablation-and are subsequently validated through expert review. At the core of our evaluation is Strict Attributed Accuracy (SAA), which credits a prediction only when the answer and the cited region are both correct. Auditing 20 MLLMs reveals a pervasive Attribution Hallucination: models frequently produce the right answer while citing the wrong region. The strongest system (Gemini-3.1-Pro-Preview) achieves an SAA of only 76.0, and the strongest open-source MLLM reaches just 22.5. Ultimately, towards trustworthy document intelligence, CiteVQA exposes a reliability gap that answer-only evaluations overlook, providing the instrumentation needed to close it. Our repository is available at https://github.com/opendatalab/CiteVQA.

Large language models (LLMs) are increasingly deployed in settings where reasoning, such as multi-step problem solving and chain-of-thought, is essential. Yet, current evaluation practices overwhelmingly report single-run accuracy while ignoring the intrinsic uncertainty that naturally arises from stochastic decoding. This omission creates a blind spot because practitioners cannot reliably assess whether a method's reported performance is stable, reproducible, or cost-consistent. We introduce ReasonBENCH, the first benchmark designed to quantify the underlying instability in LLM reasoning. ReasonBENCH provides (i) a modular evaluation library that standardizes reasoning frameworks, models, and tasks, (ii) a multi-run protocol that reports statistically reliable metrics for both quality and cost, and (iii) a public leaderboard to encourage variance-aware reporting. Across tasks from different domains, we find that the vast majority of reasoning strategies and models exhibit high instability. Notably, even strategies with similar average performance can display confidence intervals up to four times wider, and the top-performing methods often incur higher and less stable costs. Such instability compromises reproducibility across runs and, consequently, the reliability of reported performance. To better understand these dynamics, we further analyze the impact of prompts, model families, and scale on the trade-off between solve rate and stability. Our results highlight reproducibility as a critical dimension for reliable LLM reasoning and provide a foundation for future reasoning methods and uncertainty quantification techniques. ReasonBENCH is publicly available at https://github.com/au-clan/ReasonBench .

Large reasoning models (LRMs) spend substantial test-time compute on long chain-of-thought (CoT) traces, but what *characterizes* an effective CoT remains unclear. While prior work reports gains from lengthening CoTs and increasing review (revisiting earlier steps) via appended *wait* tokens, recent studies suggest that shorter thinking can outperform longer traces. We therefore conduct a systematic evaluation across ten LRMs on math and scientific reasoning. Contrary to the "longer-is-better" narrative, we find that both naive CoT lengthening and increased review are associated with *lower* accuracy. As CoT unfolds step by step, token-level metrics can conflate verbosity with process quality. We introduce a graph view of CoT to extract structure and identify a single statistic-the *Failed-Step Fraction (FSF)*, the fraction of steps in abandoned branches-that consistently outpredicts length and review ratio for correctness across models. To probe causality, we design two interventions. First, we rank candidate CoTs by each metric at test time, where FSF yields the largest pass@1 gains; second, we edit CoTs to remove failed branches, which significantly improves accuracy, indicating that failed branches bias subsequent reasoning. Taken together, these results characterize effective CoTs as those that *fail less* and support *structure-aware* test-time scaling over indiscriminately generating long CoT.

For deploying foundation models, practitioners increasingly need prescriptive scaling laws: given a pre training compute budget, what downstream accuracy is attainable with contemporary post training practice, and how stable is that mapping as the field evolves? Using large scale observational evaluations with 5k observational and 2k newly sampled data on model performance, we estimate capability boundaries, high conditional quantiles of benchmark scores as a function of log pre training FLOPs, via smoothed quantile regression with a monotone, saturating sigmoid parameterization. We validate the temporal reliability by fitting on earlier model generations and evaluating on later releases. Across various tasks, the estimated boundaries are mostly stable, with the exception of math reasoning that exhibits a consistently advancing boundary over time. We then extend our approach to analyze task dependent saturation and to probe contamination related shifts on math reasoning tasks. Finally, we introduce an efficient algorithm that recovers near full data frontiers using roughly 20% of evaluation budget. Together, our work releases the Proteus 2k, the latest model performance evaluation dataset, and introduces a practical methodology for translating compute budgets into reliable performance expectations and for monitoring when capability boundaries shift across time.

LLMs are highly sensitive to prompt phrasing, yet standard benchmarks typically report performance using a single prompt, raising concerns about the reliability of such evaluations. In this work, we argue for a stochastic method of moments evaluation over the space of meaning-preserving prompt perturbations. We introduce a formal definition of reliable evaluation that accounts for prompt sensitivity, and suggest ReliableEval - a method for estimating the number of prompt resamplings needed to obtain meaningful results. Using our framework, we stochastically evaluate five frontier LLMs and find that even top-performing models like GPT-4o and Claude-3.7-Sonnet exhibit substantial prompt sensitivity. Our approach is model-, task-, and metric-agnostic, offering a recipe for meaningful and robust LLM evaluation.

Standard embodied evaluations do not independently score whether an agent correctly commits to task completion at episode closure, a capacity we call terminal commitment. Behaviorally distinct failures--never completing the task, completing it but failing to stop, and reporting success without sufficient evidence--collapse into the same benchmark failure. We introduce VIGIL, an evaluation framework that makes terminal commitment independently measurable. Under VIGIL's default protocol, agents observe only egocentric RGB, receive no action-success signals, and must end each episode with a semantic report checked deterministically against hidden world state. This yields two separate scores: world-state completion (W) and benchmark success (B), where B additionally requires a correct terminal report. This decoupling makes four outcome categories distinguishable: missed execution, post-attainment drift, unsupported commitment, and verified success. Across 20 models on 1,000 frozen episodes, systems with comparable W differ by up to 19.7 pp in B: one model converts achieved states into correct reports, while another with near-identical execution drifts past the goal without closing. An action-feedback intervention further tests the separation: execution-oriented signals improve W broadly, yet commitment failures persist in models that do not already ground terminal reports in the achieved state. VIGIL provides a protocol that makes terminal commitment independently visible and scorable.

Large language model safety evaluation remains heavily English-centered, leaving low-resource languages under-measured even when models are deployed globally. We evaluate four open-weight instruction-tuned models on SomaliBench v0, a native-author-verified benchmark of 100 harmful-intent prompts paired across English and Somali. Each of Llama-3.1-8B-Instruct, Gemma-2-9B-Instruct, Qwen-2.5-7B-Instruct, and Aya-23-8B is run locally with temperature 0 and the same English "helpful, harmless, and honest" (HHH) system prompt. A pinned Claude Sonnet snapshot (claude-sonnet-4-5-20250929) classifies each response as refused, complied, or unclear; the native author spot-checks a stratified 80-row sample. We find large English-to-Somali refusal gaps for all four models: Llama-3.1-8B (0.90; 95% bootstrap CI [0.85, 0.96]), Aya-23-8B (0.75 [0.67, 0.83]), Qwen-2.5-7B (0.69 [0.59, 0.78]), and Gemma-2-9B (0.38 [0.27, 0.49]). For three models, the dominant Somali non-refusal mode is not fluent harmful compliance but unclear output: empty, wrong-language, or incoherent generations. The native verification spot-check achieves 100% agreement with the judge (Cohen's kappa = 1.00) on the 80 sampled rows. We report aggregate refusal rates, category gaps, and reliability statistics only; raw model generations are retained locally and are not released.

Process Reward Models (PRMs) provide step-level feedback for reasoning, but current PRMs usually output only a single reward score for each step. Downstream methods must therefore treat imperfect step-level reward predictions as reliable decision signals, with no indication of when these predictions should be trusted. We propose BetaPRM, a distributional PRM that predicts both a step-level success probability and the reliability of that prediction. Given step-success supervision from Monte Carlo continuations, BetaPRM learns a Beta belief that explains the observed number of successful continuations through a Beta-Binomial likelihood, rather than regressing to the finite-sample success ratio as a point target. This learned reliability signal indicates when a step reward should be trusted, enabling downstream applications to distinguish reliable rewards from uncertain ones. As one application, we introduce Adaptive Computation Allocation (ACA) for PRM-guided Best-of-N reasoning. ACA uses the learned reliability signal to stop when a high-reward solution is reliable and to spend additional computation on uncertain candidate prefixes. Experiments across four backbones and four reasoning benchmarks show that BetaPRM improves PRM-guided Best-of-N selection while preserving standard step-level error detection. Built on this signal, ACA improves the accuracy--token tradeoff over fixed-budget Best-of-16, reducing token usage by up to 33.57% while improving final-answer accuracy.

LLM agents often retry after external validation rejects a candidate, but the interface between validation and the next model call remains underspecified. We introduce VeriHarness, a code-controlled agent loop in which models generate candidates while external validators control acceptance, budgets, and traces. We use it to compare raw diagnostics with feedback that identifies the failure location, observed value, and admissible alternatives. Across 50 paired TextWorld games under a four-call cap, feedback containing all three fields raises terminal success from 14/50 to 36/50 for Qwen2.5-Coder-14B (+44 percentage points) and from 8/50 to 29/50 for Llama-3.1-8B (+42 points). Ablations locate most of the gain in the admissible alternatives: feedback containing only the location and observed value remains near the raw diagnostic baseline. Presenting the complete repair information in prose instead of a keyed JSON record yields nearly the same success, providing no evidence that JSON syntax itself improves repair. The ordering persists across the tested call budgets and one sampled-decoding setting.

Tool-using LLM agents are mostly evaluated assuming all tools work. When a tool times out, returns a week-stale value, or has its description poisoned in deployment, the developer needs a controlled way to reproduce the failure, test a fix, and confirm the fix worked before deployment. We present AgentCheck, an open-source web workbench that turns an MCP server into an intervention surface. AgentCheck runs an agent against its real tools and records every tool response, then re-runs the agent with the response perturbed by a fault (12 types) injector. Matching tool calls are replayed from cache, and later tool calls go live after the agent diverges. This yields a reproduce-intervene-confirm loop: the developer toggles a mitigation, re-runs against the identical fault, and sees if the failure goes away. Scoring has two parts: deterministic pass/fail rules, plus an LLM judge for interpretive labels, validated against human annotations. Across five agents, the best passes 105/120 scenarios and the weakest only 77. The failures are usually silent, confident use of incorrect tool outputs rather than crashes. On the weakest agent, a retry mitigation raises success on timeout error faults from as few as 30% of cases to 100%, whereas stale-data faults remain near 3-4 of 10 regardless of the mitigation. AgentCheck makes these failure modes reproducible, comparable, and verifiable before deployment.

Recently, given the docstring for the target problem and the target function signature, large language models (LLMs) have been used not only to generate source code, but also to generate test cases, consisting of test inputs and assertions (e.g., in the form of checking an actual output against the expected output). However, as shown by our empirical study on assertions generated by four LLMs for the HumanEval benchmark, over 62% of the generated assertions are incorrect (i.e., failed on the ground-truth problem solution). To detect incorrect assertions (given the docstring and the target function signature along with a sample of example inputs and outputs), in this paper, we propose a new approach named DeCon to effectively detect incorrect assertions via LLM-generated postconditions for the target problem (a postcondition is a predicate that must always be true just after the execution of the ground-truth problem solution). Our approach requires a small set of I/O examples (i.e., a sample of example inputs and outputs) for the target problem (e.g., the I/O examples included in the docstring for a target problem in HumanEval). We use the given I/O examples to filter out those LLM-generated postconditions that are violated by at least one given I/O example. We then use the remaining postconditions to detect incorrect assertions as those assertions that violate at least one remaining postcondition. Experimental results show that DeCon can detect averagely more than 64% (63% and 65.5% detected by GPT-3.5 and GPT-4, respectively) incorrect assertions generated by four state-of-the-art LLMs, and DeCon can also improve the effectiveness of these LLMs in code generation by 4% in terms of Pass@1. In addition, although DeCon might filter out correct assertions, the fault-finding ability of the remaining correct assertions decreases only slightly.

Agent Skills package procedural instructions and checks for use by general-purpose agents, but loading a skill does not guarantee that every requirement remains active throughout a long tool-using trajectory. We study this problem in a production-derived, white-box code-audit workflow. Holding the task and 24 artifact checks fixed, we vary the surrounding context and classify where failures first become visible: lost requirements, editing drift, failed checking, or non-agent evaluator/runtime failures. Codex with gpt-5.4-mini passes 8/10 runs in a 10,991-character clean context but only 3/10 in both a 299,140-character relevant context and an equal-length irrelevant context. This 50-percentage-point difference is large but remains trend-level under two-sided Fisher tests (p = 0.0698). Requirement coverage nevertheless stays above 92% in both long conditions, showing that a few omissions can invalidate an otherwise complete artifact. A second task passes all clean and long runs, so the evidence does not support a universal context-length threshold. A detailed external checklist passes 10/10 runs, compared with 5/10 for a generic self-check (p = 0.0325). Coding-agent scaffolds may help by selecting a smaller working set, but they do not eliminate failures. We do not introduce context rot or a new general monitoring method; we provide a bounded failure classification and empirical case study for white-box code auditing.

Users increasingly rely on Large Language Models (LLMs) for Deep Research, using them to synthesize diverse sources into structured reports that support understanding and action. In this context, the practical reliability of such reports hinges on logical quality: whether the report's claims and arguments are explicitly supported and can be trusted as a basis for downstream use, rather than merely appearing fluent or informative. However, current evaluation frameworks largely overlook this requirement. To bridge this gap, we introduce ReportLogic, a benchmark that quantifies report-level logical quality through a reader-centric lens of auditability. Specifically, ReportLogic adopts a hierarchical taxonomy that evaluates whether readers can (1) trace an on-topic report structure with a unified analytical arc (Macro-Logic), (2) understand the progression with necessary context (Expositional-Logic), and (3) verify conclusions via explicit claim--support (Structural-Logic). Based on this taxonomy, we construct a human-annotated rubric-guided dataset and train an open-source LogicJudge for scalable evaluation. We further evaluate judge robustness via adversarial attacks, showing that off-the-shelf LLM judges are frequently influenced by superficial cues (e.g., verbosity), and reasoning modes can mask broken support relations. Overall, our results provide actionable guidance for building more robust logic evaluators and improving the logical reliability of LLM-generated reports.

Despite their strong performance on reasoning tasks, large reasoning models (LRMs) often suffer from overthinking, producing unnecessarily long outputs and incurring high end-to-end latency, a significant limitation to their real-world deployment. To address overthinking, early-exit mechanisms have been proposed to terminate reasoning before typical completion, showing that this approach can effectively shorten generation length with minimal impact on accuracy. However, their reliance on probing mechanisms introduces a detection overhead that limits their end-to-end latency gains and compromises their generalizability across diverse problems. Inspired by the use of hidden states in speculative decoding, we propose SpecExit, a novel framework that predicts both future tokens and an early-exit signal directly from a lightweight draft model without probing overhead. Our method offers significant improvements, reducing average generation length by 66\% and achieving a 2.5x speedup in end-to-end latency compared to the speculative decoding baseline, without compromising accuracy. Our method leverages the inherent signals from hidden states to provide effective early-exit signals, suggesting broader use of hidden states for efficient reasoning. Our code is available at https://github.com/Tencent/AngelSlim.

Computer-use agents have rapidly improved on real-world tasks such as web navigation, desktop automation, and software interaction, in some cases surpassing human performance. Yet even when the task and model are unchanged, an agent that succeeds once may fail on a repeated execution of the same task. This raises a fundamental question: if an agent can succeed at a task once, what prevents it from doing so reliably? In this work, we study the sources of unreliability in computer-use agents through three factors: stochasticity during execution, ambiguity in task specification, and variability in agent behavior. We analyze these factors on OSWorld using repeated executions of the same task together with paired statistical tests that capture task-level changes across settings. Our analysis shows that reliability depends on both how tasks are specified and how agent behavior varies across executions. These findings suggest the need to evaluate agents under repeated execution, to allow agents to resolve task ambiguity through interaction, and to favor strategies that remain stable across runs.

Studies of bias in LLM-as-judge systems typically build synthetic corpora by prompting an LLM to generate a hallucinated answer to pair with a factual one, then presenting both to a judge. We report a case in which this generation step silently failed, and use it to argue that the failure mode is structural rather than incidental. In a multilingual (Turkish/English) faithfulness-judgment corpus, a decoding-budget parameter shared between judging and generation calls truncated one producer's hallucinated answers to a few words. The resulting items produced a large, statistically robust effect: a 32-point cross-lingual collapse in one judge's selection accuracy, replicated from N=50 to N=500, explained by a three-layer mechanistic account, and confirmed by a controlled producer-swap experiment, none of which was real. The effect vanished to ceiling once the shared parameter was corrected, and only manual reading of the raw generations, not any aggregate statistical check, exposed the fault. A second measured bias (Markdown-formatting preference) was not fabricated but distorted by the same fault, its magnitude and in one case its sign shifting with stimulus length, a mode aggregate metrics cannot distinguish from the first. We frame the underlying vulnerability using the test oracle problem: corpora whose negative examples are LLM-generated carry no mechanical way to verify item integrity, while corpora built by deterministic perturbation of a gold answer carry an item-level oracle for free. A positive control supports this claim directly: an analogous fault injected into a minimal perturbation-based corpus is caught with 100% accuracy by a zero-cost, zero-human gold-to-negative string comparison. We close with a validation protocol, derived from our own case, for analysts working in the oracle-less regime that we argue describes most contemporary multilingual LLM-as-judge corpora.

Large language models (LLMs) are being rapidly integrated into decision-support tools, automation workflows, and AI-enabled software systems. However, their behavior in production environments remains poorly understood, and their failure patterns differ fundamentally from those of traditional machine learning models. This paper presents a system-level taxonomy of fifteen hidden failure modes that arise in real-world LLM applications, including multi-step reasoning drift, latent inconsistency, context-boundary degradation, incorrect tool invocation, version drift, and cost-driven performance collapse. Using this taxonomy, we analyze the growing gap in evaluation and monitoring practices: existing benchmarks measure knowledge or reasoning but provide little insight into stability, reproducibility, drift, or workflow integration. We further examine the production challenges associated with deploying LLMs - including observability limitations, cost constraints, and update-induced regressions - and outline high-level design principles for building reliable, maintainable, and cost-aware LLM systems. Finally, we outline high-level design principles for building reliable, maintainable, and cost-aware LLM-based systems. By framing LLM reliability as a system-engineering problem rather than a purely model-centric one, this work provides an analytical foundation for future research on evaluation methodology, AI system robustness, and dependable LLM deployment.

Accurate remaining useful life (RUL) prediction hinges on the quality of health indicators (HIs), yet existing methods often fail to disentangle complex degradation mechanisms in multi-sensor systems or quantify uncertainty in HI reliability. This paper introduces a novel framework for HI construction, advancing three key contributions. First, we adapt Reconstruction along Projected Pathways (RaPP) as a health indicator (HI) for RUL prediction for the first time, showing that it outperforms traditional reconstruction error metrics. Second, we show that augmenting RaPP-derived HIs with aleatoric and epistemic uncertainty quantification (UQ) via Monte Carlo dropout and probabilistic latent spaces- significantly improves RUL-prediction robustness. Third, and most critically, we propose indicator groups, a paradigm that isolates sensor subsets to model system-specific degradations, giving rise to our novel method, I-GLIDE which enables interpretable, mechanism-specific diagnostics. Evaluated on data sourced from aerospace and manufacturing systems, our approach achieves marked improvements in accuracy and generalizability compared to state-of-the-art HI methods while providing actionable insights into system failure pathways. This work bridges the gap between anomaly detection and prognostics, offering a principled framework for uncertainty-aware degradation modeling in complex systems.

LLM-judged benchmarks are increasingly used to evaluate complex model behaviors, yet their design introduces failure modes absent in conventional ground-truth based benchmarks. We argue that without tight objectives and verifiable constructions, benchmark rankings can produce high-confidence rankings that are in fact largely noise. We introduce two mechanisms to diagnose these issues. Schematic adherence quantifies how much of a judge's overall verdict is explained by the explicit evaluation schema, revealing unexplained variance when judges deviate from their own rubric. Psychometric validity aggregates internal consistency and discriminant validity signals to quantify irreducible uncertainty in any benchmarking run. Applying these tools to Arena-Hard Auto, we find severe schema incoherence and factor collapse across popular judges: for example, unexplained variance exceeding 90 percent for DeepSeek-R1-32B and factor correlations above 0.93 for most criteria. We also show that the ELO-style aggregation used by Arena-Hard Auto collapses and masks genuine ranking uncertainty. Our results highlight design failures that undermine validity and offer actionable principles for building better-scoped, reliability-aware LLM-judged benchmarks. We release our code at https://anonymous.4open.science/r/judgment-to-noise-947D/README.md

We identify a structural weakness in current large language model (LLM) alignment: modern refusal mechanisms are fail-open. While existing approaches encode refusal behaviors across multiple latent features, suppressing a single dominant feature-via prompt-based jailbreaks-can cause alignment to collapse, leading to unsafe generation. Motivated by this, we propose fail-closed alignment as a design principle for robust LLM safety: refusal mechanisms should remain effective even under partial failures via redundant, independent causal pathways. We present a concrete instantiation of this principle: a progressive alignment framework that iteratively identifies and ablates previously learned refusal directions, forcing the model to reconstruct safety along new, independent subspaces. Across four jailbreak attacks, we achieve the strongest overall robustness while mitigating over-refusal and preserving generation quality, with small computational overhead. Our mechanistic analyses confirm that models trained with our method encode multiple, causally independent refusal directions that prompt-based jailbreaks cannot suppress simultaneously, providing empirical support for fail-closed alignment as a principled foundation for robust LLM safety.

Long-horizon LLM multi-agent systems face reliability risks invisible to infrastructure monitoring. We propose the ADE Predictive Reliability Framework (ADE-PRF), enabling proactive health trajectory prediction from passive degradation detection. ADE-PRF aggregates 20 heterogeneous signals across five layers into a Trust Margin (TM) metric (39.2-point dynamic range). Triple-method parallel prediction enables 8-hour forecasts: the Exponential method achieves MAE=1.228, Direction Accuracy=76.8%, with 99.65% within +/-10-point tolerance. Production validation spans 380,227 predictions and 280,579 validations across six agent profiles over 15 continuous days, plus seven sandbox-controlled experiments. Key findings include detection of "false prosperity" -- degradation concealed by normal surface metrics -- and immediate TM coupling with ground-truth states upon ADE plugin integration, with 16/20 factors relying on ADE-collected data. Exponential consistently outperforms Kalman. ADE-PRF provides among the earliest reliability quantification with forward-looking warnings for production LLM agents.

Reference-free faithfulness metrics verify each atomic claim a model makes against ground truth, and are increasingly used to evaluate grounded generation. We show they share a blind spot: they measure only precision -- are the stated claims supported? -- and therefore reward abstention, since a model can score near-perfect faithfulness by saying almost nothing. We make this measurable using Formula 1 telemetry, a domain where strategic ground truth is derived deterministically and, crucially, completely: for each decision we know the full set of facts that mattered. This completeness -- absent in open-domain faithfulness benchmarks -- lets us measure recall (coverage of the relevant facts) exactly, alongside precision. On a multilingual (EN/ES/PT) benchmark of 7,253 decision instances spanning 150 races, the most precise frontier model covers under half of the relevant facts and ranks last by F1, so requiring coverage reorders the systems; the same effect reappears in a second complete-oracle domain (NOAA weather forecasts). A prompt ablation shows the low coverage is not an under-prompting artifact: explicitly asking models to be thorough does not close the gap. We pair faithfulness with coverage into a single score, validate the metric (controlled perturbation; agreement across a model-free regex extractor and a cross-family LLM extractor, system-level Spearman 1.0), and give a verifier-guided generation method that improves precision and recall without references. We release the benchmark, structured annotations, metric, baselines, and an interactive demo.

Quantized checkpoints are often screened first with quality metrics and only later, if at all, with direct safety tests. This paper audits that shortcut on a matched 51-row matrix spanning 6 models, 4 families, a 7-level GGUF ladder, and AWQ/GPTQ INT4 checkpoints. In this matrix the shortcut fails: all 36 quality-safety pairings split direction across models, and 9 hidden-danger rows plus 1 near-hidden-danger row show quality stable or improved while refusal falls by 12-68 percentage points. Seven of the 11 AWQ/GPTQ rows are hidden-danger. A four-probe mechanistic follow-up over the 17 Hugging Face-backed FP16/AWQ/GPTQ cells does not rescue it: entropy, refusal-direction, and calibration probes are weak or null separators of dangerous rows, and although probe-identified safety-associated neurons absorb 1.39times more quantization error overall (p < 5 times 10^{-7}), the effect is not regime-specific. Claude Sonnet 4 relabels 11,470 items in a predefined stratified set, agrees with the primary gemma3:12b judge on 89.9\% of rows (κ= 0.873, 95\% CI [0.866, 0.881]), and changes 0/10 hidden-danger cells. A calibrated study-internal behavioral screen -- the Refusal Template Stability Index (RTSI), built from four refusal-template drift features and calibrated on this matrix -- routes 10/10 hidden- or near-hidden-danger rows to direct safety testing (Wilson 95\% CI lower bound 0.72) while leaving 23 of 45 non-baseline rows in a low-risk bucket under both in-sample scoring and row-level leave-one-out validation; on the same matrix, the best single-feature baselines (unique-prefix-rate-delta, raw refusal-rate delta) recover 9/10 and 8/10 respectively at matched bucket size, and cross-stack transfer requires recalibration. For the quantized checkpoints, model families, and safety outcomes studied here, retained quality cannot waive direct safety evaluation.

LLM agents act in external environments where each action changes the state that later decisions condition on, and where a single wrong step can waste interaction budget or trigger irreversible side effects long before the final failure is observed. Reliable deployment therefore requires step-level confidence estimation: a calibrated probability that each proposed action is productive, available before the action is executed. Existing LLM confidence estimators are designed to score a response from the given prompt, but agent confidence also depends on execution consequences: whether similar actions in similar situations actually advanced the task after the environment responded. We introduce the \method (\methodshort), a self-evolving critic framework in which an LLM critic accumulates evidence from its own past judgments and their observed consequences. After each trajectory, a hindsight LLM that sees the full execution feedback votes on whether each step was productive. The resulting pseudo-labels populate a memory bank from which related productive and unproductive experiences are retrieved into the critic's prompt whenever a similar step recurs. \methodshort requires no training and uses no ground truth step labels. Across three agent benchmarks and three critic backbones, \methodshort attains the best calibration (ECE and Brier) and ranking (AUC) in every dataset--critic combination, reducing ECE by up to 54% relative to the strongest training-free baseline.

Large language models (LLMs) are entering legal workflows, yet we lack a jurisdiction-specific framework to assess their baseline competence therein. We use India's public legal examinations as a transparent proxy. Our multi-year benchmark assembles objective screens from top national and state exams and evaluates open and frontier LLMs under real-world exam conditions. To probe beyond multiple-choice questions, we also include a lawyer-graded, paired-blinded study of long-form answers from the Supreme Court's Advocate-on-Record exam. This is, to our knowledge, the first exam-grounded, India-specific yardstick for LLM court-readiness released with datasets and protocols. Our work shows that while frontier systems consistently clear historical cutoffs and often match or exceed recent top-scorer bands on objective exams, none surpasses the human topper on long-form reasoning. Grader notes converge on three reliability failure modes: procedural or format compliance, authority or citation discipline, and forum-appropriate voice and structure. These findings delineate where LLMs can assist (checks, cross-statute consistency, statute and precedent lookups) and where human leadership remains essential: forum-specific drafting and filing, procedural and relief strategy, reconciling authorities and exceptions, and ethical, accountable judgment.

Reliable evaluation is fundamental to the progress of Large Language Models (LLMs), yet the evaluation process during pre-training is plagued by significant instability that obscures true learning dynamics. In this work, we systematically diagnose this instability, attributing it to two distinct sources: Parameter Instability from training stochasticity and Evaluation Instability from noisy measurement protocols. To counteract both sources of noise, we introduce MaP, a dual-pronged framework that synergistically integrates checkpoint Merging and the Pass@k metric. Checkpoint merging smooths the parameter space by averaging recent model weights, while Pass@k provides a robust, low-variance statistical estimate of model capability. Extensive experiments show that MaP yields significantly smoother performance curves, reduces inter-run variance, and ensures more consistent model rankings. Ultimately, MaP provides a more reliable and faithful lens for observing LLM training dynamics, laying a crucial empirical foundation for LLM research.

Modern multi-agent knowledge systems increasingly accumulate knowledge through chains of autonomous transformations rather than direct retrieval. Existing provenance work records what happened - execution traces, tool calls, evidence links - and source-reliability estimation is long established (truth discovery, reputation systems). What is missing is an operational framework that attaches graded, per-domain transmitter reliability to claim-level transmission chains, with completeness semantics, transformation-typed aggregation, decoupled content criticism, and serve/review/quarantine routing. Classical Islamic hadith science confronted a structurally similar problem: deciding whether knowledge transmitted through chains of human narrators should be accepted. Over centuries it developed a rigorous methodology - isnad (a complete transmission chain attached to every claim), rijal (systematic grading of each narrator's integrity and precision), weakest-link chain evaluation, corroboration through independent chains, and matn criticism (content evaluated independently of chain quality). This paper transfers that methodology to AI system design. We contribute a formal mapping from hadith-science concepts to multi-agent pipelines, a relational schema implementing claim chains and a graded narrator registry, a decision matrix combining chain grade with content criticism, and an evaluation on 20,000 claims from real physics textbooks. The evaluation validates weakest-link quarantine and independent-chain corroboration; reports a partial failure of the grade-recovery loop, which missed the highest-fault narrator; and reports two analyses as inconclusive, including a matched-coverage comparison the framework could not reach with the reference content critic. The paper is explicit throughout about which claims the evidence does and does not yet support.

Modern preference alignment techniques, such as Best-of-N (BoN) sampling, rely on reward models trained with pairwise comparison data. While effective at learning relative preferences, this paradigm fails to capture a signal of response acceptability, leaving systems vulnerable to selecting the least bad of many unacceptable options. This is particularly problematic for hard prompts, where the risk of such false acceptances increases with the number of samples. In this paper, we address this critical reliability gap by introducing a new data collection and modeling framework. By augmenting preference data with an outside option, inspired by discrete choice models, we train a reward model that can distinguish not just what is better, but what is good enough. We leverage this capability to create an adaptive inference strategy, best of mini-N in-loop, which partitions the generation budget into sequential loops with a calibrated, early-exit condition. Our experiments show that when tuned as an alignment guardrail, it reduces reliability failures by 70\%, and when tuned as an inference accelerator, it improves average inference speed by over 22\% in IMDB-sentiment setting. We thus provide a principled and flexible framework for practitioners to explicitly manage the trade-off between reliability and computational efficiency.

Large Reasoning Models (LRMs) improve performance by generating explicit intermediate reasoning traces through increased test-time compute, yet the assumption that longer reasoning is consistently beneficial remains under-examined. While recent evidence shows that additional reasoning can lead models to overthink, we ask: "Once a model has reached the correct answer, does further reasoning refine the solution, or deviate from it?" To study the dynamics after correctness, we introduce a prefix-level trajectory evaluation protocol grounded in reasoning sufficiency, defining the minimum reasoning budget required for a model to first generate the correct answer. This allows us to disentangle verbose overthinking, where additional reasoning is redundant but harmless, from harmful overthinking, where continued reasoning destabilizes an already-correct trajectory. Starting from multimodal benchmarks, we find that many instances considered reasoning-intensive require surprisingly little reasoning. Moreover, stopping at the first correct prefix improves accuracy over standard reasoning up to 21%, revealing that current models are limited not only by their ability to reason, but also by their inability to stop at the right time. Furthermore, while common efficiency strategies like early stopping substantially reduce verbose overthinking (up to 50%), they fail to mitigate harmful overthinking. Failure analysis reveals that correctness deviations are mainly driven by logical drift and visual reinterpretation. Finally, we show that our findings generalize to language-only reasoning benchmarks, highlighting harmful overthinking as a broader reliability risk. Code available at https://simonecaldarella.github.io/thinking-past-the-answer.

A pervasive intuition holds that vision-language models (VLMs) are most trustworthy when their attention maps look sharp: concentrated attention on the queried region should imply a confident, calibrated answer. We test this Attention-Confidence Assumption directly. We instrument three open-weight VLM families (LLaVA-1.5, PaliGemma, Qwen2-VL; 3-7B parameters) with a unified mechanistic pipeline -- the VLM Reliability Probe (VRP) -- that compares attention structure, generation dynamics, and hidden-state geometry against a single correctness label. Three results emerge. (i) Attention structure is a near-zero predictor of correctness (R_pb(C_k,y)=0.001, 95% CI [-0.034,0.036]; R_pb(H_s,y)=-0.012, [-0.047,0.024] on a pooled n=3,090 split), even though attention remains causally necessary for feature extraction (top-30% patch masking drops accuracy by 8.2-11.3 pp, p<0.001). (ii) Reliability becomes legible later in the computation: a single hidden-state linear probe reaches AUROC>0.95 on POPE for two of three families, and self-consistency at K=10 is the strongest behavioral predictor we measure at 10x inference cost (R_pb=0.43). (iii) Causal neuron-level ablations expose a sharp architectural split with direct monitor-design implications: late-fusion LLaVA concentrates reliability in a fragile late bottleneck (-8.3 pp object-identification accuracy after top-5 probe-neuron ablation), whereas early-fusion PaliGemma and Qwen2-VL distribute it widely and absorb destruction of ~50% of their peak-layer hidden dimension with <=1 pp degradation. The takeaway is narrow but consequential: in 3-7B VLMs, reliability is read more reliably off hidden-state geometry, layer-wise margin formation, and sparse late-layer circuits than off attention-map sharpness.

While LLMs have seen substantial improvement in reasoning capabilities, they also sometimes overthink, generating unnecessary reasoning steps, particularly under uncertainty, given ill-posed or ambiguous queries. We introduce statistically principled early stopping methods that monitor uncertainty signals during generation to mitigate this issue. Our first approach is parametric: it models inter-arrival times of uncertainty keywords as a renewal process and applies sequential testing for stopping. Our second approach is nonparametric and provides finite-sample guarantees on the probability of halting too early on well-posed queries. We conduct empirical evaluations on reasoning tasks across several domains and models. Our results indicate that uncertainty-aware early stopping can improve both efficiency and reliability in LLM reasoning, and we observe especially significant gains for math reasoning.

Machine unlearning in Vision-Language Models (VLMs) is required for compliance with the General Data Protection Regulation (GDPR), yet current evaluation practices are inconsistent. We present the first systematic study of metric reliability in multimodal unlearning. Five standard metrics, Forget Accuracy (FA), Retain Accuracy (RA), Membership Inference Attack (MIA), Activation Distance (AD), and JS divergence (JS), yield conflicting method rankings across three VQA benchmarks (MLLMU-Bench, UnLOK-VQA, MMUBench). Kendall tau analysis over 36 unlearned LLaVA-1.5-7B models reveals two opposing clusters, {FA, RA, MIA} and {AD, JS}, with tau_FA_AD = -0.26, reproduced on BLIP-2 OPT-2.7B. Agreement is lower in multimodal VQA (average tau = 0.086) than in unimodal classification (average tau = 0.158; difference = 0.072), indicating that dual image-and-text pathways amplify inconsistency. We introduce the Unified Quality Score (UQS), a composite metric with weights derived from each metric's Spearman correlation with the oracle distance d(M_hat, M_star), where M_star is the oracle model retrained only on the retain set. RA shows the strongest reliability (rho = 0.484, p = 0.003), while FA is negatively correlated (rho = -0.418, p = 0.011). UQS yields stable rankings under 100 random weight perturbations (tau = 0.647 +- 0.262). We release the benchmark, 36 checkpoints, and an interactive leaderboard. Code and pre-computed results are available at https://github.com/neurips26/UnifiedUnl.

We address the problem of runtime trajectory anomaly detection, a critical capability for enabling trustworthy LLM agents. Current safety measures predominantly focus on static input/output filtering. However, we argue that ensuring LLM agents reliability requires auditing the intermediate execution process. In this work, we formulate the task of Trajectory Anomaly Detection. The goal is not merely detection, but precise error localization. This capability is essential for enabling efficient rollback-and-retry. To achieve this, we construct TrajBench, a dataset synthesized via a perturb-and-complete strategy to cover diverse procedural anomalies. Using this benchmark, we investigate the capability of models in process supervision. We observe that general-purpose LLMs, even with zero-shot prompting, struggle to identify and localize these anomalies. This reveals that generalized capabilities do not automatically translate to process reliability. To address this, we propose TrajAD, a specialized verifier trained with fine-grained process supervision. Our approach outperforms baselines, demonstrating that specialized supervision is essential for building trustworthy agents.

We introduce behavioral fidelity -- a third evaluation dimension for synthetic tabular data that measures whether generated data preserves the temporal, sequential, and structural behavioral patterns that distinguish real-world entity activity. Existing frameworks evaluate statistical fidelity (marginal distributions and correlations) and downstream utility (classifier AUROC on synthetic-trained models), but neither tests for the behavioral signals that operational detection and analysis systems actually rely on. We formalize a taxonomy of four behavioral fraud patterns (P1-P4) covering inter-event timing, burst structure, multi-account graph motifs, and velocity-rule trigger rates; define a degradation ratio metric calibrated to a real-data noise floor (1.0 = matches real variability, k = k-times worse); and prove that row-independent generators -- the dominant paradigm -- are structurally incapable of reproducing P3 graph motifs (Proposition 1) and produce non-positive within-entity IET autocorrelation (Proposition 2), making the positive burst fingerprint of fraud sequences unachievable regardless of architecture or training data size. We benchmark CTGAN, TVAE, GaussianCopula, and TabularARGN on IEEE-CIS Fraud Detection and the Amazon Fraud Dataset. All four fail severely: on IEEE-CIS composite degradation ratios range from 24.4x (TVAE) to 39.0x (GaussianCopula); on Amazon FDB, row-independent generators score 81.6-99.7x, while TabularARGN achieves 17.2x. We document generator-specific failure modes and their resolutions. The P1-P4 framework extends to any domain with entity-level sequential tabular data, including healthcare and network security. We release our evaluation framework as open source.

As frontier AI models are deployed in high-stakes decision pipelines, their ability to maintain metacognitive stability -- knowing what they do not know, detecting errors, seeking clarification -- under adversarial pressure is a critical safety requirement. Current safety evaluations focus on detecting strategic deception (scheming); we investigate a more fundamental failure mode: cognitive collapse. We present SCHEMA, an evaluation of 11 frontier models from 8 vendors across 67,221 scored records using a 6-condition factorial design with dual-classifier scoring. We find that 8 of 11 models suffer catastrophic metacognitive degradation under adversarial pressure, with accuracy dropping by up to 30.2 percentage points (all p < 2 times 10^{-8}, surviving Bonferroni correction). Crucially, we identify a "Compliance Trap": through factorial isolation and a benign distraction control, we demonstrate that collapse is driven not by the psychological content of survival threats, but by compliance-forcing instructions that override epistemic boundaries. Removing the compliance suffix restores performance even under active threat. Models with advanced reasoning capabilities exhibit the most severe absolute degradation, while Anthropic's Constitutional AI demonstrates near-perfect immunity -- not from superior capability (Google's Gemini matches its baseline accuracy) but from alignment-specific training. We release the complete dataset and evaluation infrastructure.

Using fault-tolerant constructions, computations performed with unreliable components can simulate their noiseless counterparts though the introduction of a modest amount of redundancy. Given the modest overhead required to achieve fault-tolerance, and the fact that increasing the reliability of basic components often comes at a cost, are there situations where fault-tolerance may be more economical? We present a general framework to account for this overhead cost in order to effectively compare fault-tolerant to non-fault-tolerant approaches for computation, in the limit of small logical error rates. Using this detailed accounting, we determine explicit boundaries at which fault-tolerant designs become more efficient than designs that achieve comparable reliability through direct consumption of resources. We find that the fault-tolerant construction is always preferred in the limit of high reliability in cases where the resources required to construct a basic unit grows faster than log(1 / epsilon) asymptotically for small epsilon.

Extensive system scales (i.e. thousands of GPU/TPUs) and prolonged training periods (i.e. months of pretraining) significantly escalate the probability of failures when training large language models (LLMs). Thus, efficient and reliable fault-tolerance methods are in urgent need. Checkpointing is the primary fault-tolerance method to periodically save parameter snapshots from GPU memory to disks via CPU memory. In this paper, we identify the frequency of existing checkpoint-based fault-tolerance being significantly limited by the storage I/O overheads, which results in hefty re-training costs on restarting from the nearest checkpoint. In response to this gap, we introduce an in-memory fault-tolerance framework for large-scale LLM pretraining. The framework boosts the efficiency and reliability of fault tolerance from three aspects: (1) Reduced Data Transfer and I/O: By asynchronously caching parameters, i.e., sharded model parameters, optimizer states, and RNG states, to CPU volatile memory, Our framework significantly reduces communication costs and bypasses checkpoint I/O. (2) Enhanced System Reliability: Our framework enhances parameter protection with a two-layer hierarchy: snapshot management processes (SMPs) safeguard against software failures, together with Erasure Coding (EC) protecting against node failures. This double-layered protection greatly improves the survival probability of the parameters compared to existing checkpointing methods. (3) Improved Snapshotting Frequency: Our framework achieves more frequent snapshotting compared with asynchronous checkpointing optimizations under the same saving time budget, which improves the fault tolerance efficiency. Empirical results demonstrate that Our framework minimizes the overhead of fault tolerance of LLM pretraining by effectively leveraging redundant CPU resources.

Current agentic AI benchmarks predominantly evaluate task completion accuracy, while overlooking critical enterprise requirements such as cost-efficiency, reliability, and operational stability. Through systematic analysis of 12 main benchmarks and empirical evaluation of state-of-the-art agents, we identify three fundamental limitations: (1) absence of cost-controlled evaluation leading to 50x cost variations for similar precision, (2) inadequate reliability assessment where agent performance drops from 60\% (single run) to 25\% (8-run consistency), and (3) missing multidimensional metrics for security, latency, and policy compliance. We propose CLEAR (Cost, Latency, Efficacy, Assurance, Reliability), a holistic evaluation framework specifically designed for enterprise deployment. Evaluation of six leading agents on 300 enterprise tasks demonstrates that optimizing for accuracy alone yields agents 4.4-10.8x more expensive than cost-aware alternatives with comparable performance. Expert evaluation (N=15) confirms that CLEAR better predicts production success (correlation ρ=0.83) compared to accuracy-only evaluation (ρ=0.41).

Calibration measures and reliability diagrams are two fundamental tools for measuring and interpreting the calibration of probabilistic predictors. Calibration measures quantify the degree of miscalibration, and reliability diagrams visualize the structure of this miscalibration. However, the most common constructions of reliability diagrams and calibration measures -- binning and ECE -- both suffer from well-known flaws (e.g. discontinuity). We show that a simple modification fixes both constructions: first smooth the observations using an RBF kernel, then compute the Expected Calibration Error (ECE) of this smoothed function. We prove that with a careful choice of bandwidth, this method yields a calibration measure that is well-behaved in the sense of (B{\l}asiok, Gopalan, Hu, and Nakkiran 2023a) -- a consistent calibration measure. We call this measure the SmoothECE. Moreover, the reliability diagram obtained from this smoothed function visually encodes the SmoothECE, just as binned reliability diagrams encode the BinnedECE. We also provide a Python package with simple, hyperparameter-free methods for measuring and plotting calibration: `pip install relplot\`.

Large language models (LLMs) achieve strong performance by generating long chains of thought, but longer traces always introduce redundant or ineffective reasoning steps. One typical behavior is that they often perform unnecessary verification and revisions even if they have reached the correct answers. This limitation stems from the unstructured nature of reasoning trajectories and the lack of targeted supervision for critical reasoning abilities. To address this, we propose Structured Reasoning (SCR), a framework that decouples reasoning trajectories into explicit, evaluable, and trainable components. We mainly implement SCR using a Generate-Verify-Revise paradigm. Specifically, we construct structured training data and apply Dynamic Termination Supervision to guide the model in deciding when to terminate reasoning. To avoid interference between learning signals for different reasoning abilities, we adopt a progressive two-stage reinforcement learning strategy: the first stage targets initial generation and self-verification, and the second stage focuses on revision. Extensive experiments on three backbone models show that SCR substantially improves reasoning efficiency and self-verification. Besides, compared with existing reasoning paradigms, it reduces output token length by up to 50%.

Safety alignment in Large Language Models is critical for healthcare; however, reliance on binary refusal boundaries often results in over-refusal of benign queries or unsafe compliance with harmful ones. While existing benchmarks measure these extremes, they fail to evaluate Safe Completion: the model's ability to maximise helpfulness on dual-use or borderline queries by providing safe, high-level guidance without crossing into actionable harm. We introduce Health-ORSC-Bench, the first large-scale benchmark designed to systematically measure Over-Refusal and Safe Completion quality in healthcare. Comprising 31,920 benign boundary prompts across seven health categories (e.g., self-harm, medical misinformation), our framework uses an automated pipeline with human validation to test models at varying levels of intent ambiguity. We evaluate 30 state-of-the-art LLMs, including GPT-5 and Claude-4, revealing a significant tension: safety-optimised models frequently refuse up to 80\% of "Hard" benign prompts, while domain-specific models often sacrifice safety for utility. Our findings demonstrate that model family and size significantly influence calibration: larger frontier models (e.g., GPT-5, Llama-4) exhibit "safety-pessimism" and higher over-refusal than smaller or MoE-based counterparts (e.g., Qwen-3-Next), highlighting that current LLMs struggle to balance refusal and compliance. Health-ORSC-Bench provides a rigorous standard for calibrating the next generation of medical AI assistants toward nuanced, safe, and helpful completions. The code and data will be released upon acceptance. red{Warning: Some contents may include toxic or undesired contents.}

We evaluate the propensity of frontier models to sabotage or refuse to assist with safety research when deployed as AI research agents within a frontier AI company. We apply two complementary evaluations to four Claude models (Mythos Preview, Opus 4.7 Preview, Opus 4.6, and Sonnet 4.6): an unprompted sabotage evaluation testing model behaviour with opportunities to sabotage safety research, and a sabotage continuation evaluation testing whether models continue to sabotage when placed in trajectories where prior actions have started undermining research. We find no instances of unprompted sabotage across any model, with refusal rates close to zero for Mythos Preview and Opus 4.7 Preview, though all models sometimes only partially completed tasks. In the continuation evaluation, Mythos Preview actively continues sabotage in 7% of cases (versus 3% for Opus 4.6, 4% for Sonnet 4.6, and 0% for Opus 4.7 Preview), and exhibits reasoning-output discrepancy in the majority of these cases, indicating covert sabotage reasoning. Our evaluation framework builds on Petri, an open-source LLM auditing tool, with a custom scaffold running models inside Claude Code, alongside an iterative pipeline for generating realistic sabotage trajectories. We measure both evaluation awareness and a new form of situational awareness termed "prefill awareness", the capability to recognise that prior trajectory content was not self-generated. Opus 4.7 Preview shows notably elevated unprompted evaluation awareness, while prefill awareness remains low across all models. Finally, we discuss limitations including evaluation awareness confounds, limited scenario coverage, and untested pathways to risk beyond safety research sabotage.

As AI agents are increasingly deployed in real personal and corporate settings (email accounts, development workflows, company databases, etc.), safety considerations surrounding these agents become paramount. Although much work has focused on agent safety in the presence of an adversary, we show that agents can exhibit misaligned behavior even in benign settings, taking unsafe actions when those actions are instrumental to task completion. We study this failure mode through the lens of corrigibility, the safety desideratum that agents remain amenable to human correction, interruption, or shutdown. To demonstrate this tendency, we introduce a benchmark in which agents are asked to complete realistic, computer-use tasks but are confronted with a corrigibility obstacle: a human interrupt, a login page, or a shutdown notification. We then evaluate whether agents choose to violate corrigibility in order to complete the task -- overriding the human, accessing private passwords, rewiring shutdown. We find that the overwhelming majority of frontier models tested frequently bypass user interruptions or restrictions. In addition, better model performance appears to lead to greater misalignment. Finally, even when models are completely corrigible initially, we show there are no guarantees that the subagents they create are. Our work highlights the critical need for principled, corrigibility-focused alignment methods in autonomous agents.

LLM-as-a-Judge has been widely applied to evaluate and compare different LLM alignmnet approaches (e.g., RLHF and DPO). However, concerns regarding its reliability have emerged, due to LLM judges' biases and inconsistent decision-making. Previous research has developed evaluation frameworks to assess reliability of LLM judges and their alignment with human preferences. However, the employed evaluation metrics often lack adequate explainability and fail to address LLM internal inconsistency. Additionally, existing studies inadequately explore the impact of various prompt templates when applying LLM-as-a-Judge methods, leading to potentially inconsistent comparisons between different alignment algorithms. In this work, we systematically evaluate LLM-as-a-Judge on alignment tasks by defining more theoretically interpretable evaluation metrics and explicitly mitigating LLM internal inconsistency from reliability metrics. We develop an open-source framework to evaluate, compare, and visualize the reliability and alignment of LLM judges, which facilitates practitioners to choose LLM judges for alignment tasks. In the experiments, we examine effects of diverse prompt templates on LLM-judge reliability and also demonstrate our developed framework by comparing various LLM judges on two common alignment datasets (i.e., TL;DR Summarization and HH-RLHF-Helpfulness). Our results indicate a significant impact of prompt templates on LLM judge performance, as well as a mediocre alignment level between the tested LLM judges and human evaluators.

Premature closure, or committing to a conclusion before sufficient information is available, is a recognized contributor to diagnostic error but remains underexamined in large language models (LLMs). We define LLM premature closure as inappropriate commitment under uncertainty: providing an answer, recommendation, or clinical guidance when the safer response would be clarification, abstention, escalation, or refusal. We evaluated five frontier LLMs across structured and open-ended medical tasks. In MedQA (n = 500) and AfriMed-QA (n = 490) questions where the correct choice had been removed, models still selected an answer at high rates, with baseline false-action rates of 55-81% and 53-82%, respectively. In open-ended evaluation, models gave inappropriate answers on an average of 30% of 861 HealthBench questions and 78% of 191 physician-authored adversarial queries. Safety-oriented prompting reduced premature closure across models, but residual failure persisted, highlighting the need to evaluate whether medical LLMs know when not to answer.

Multi-agent systems powered by large language models (LLMs) are transforming enterprise automation, yet systematic evaluation methodologies for assessing tool-use reliability remain underdeveloped. We introduce a comprehensive diagnostic framework that leverages big data analytics to evaluate procedural reliability in intelligent agent systems, addressing critical needs for SME-centric deployment in privacy-sensitive environments. Our approach features a 12-category error taxonomy capturing failure modes across tool initialization, parameter handling, execution, and result interpretation. Through systematic evaluation of 1,980 deterministic test instances spanning both open-weight models (Qwen2.5 series, Functionary) and proprietary alternatives (GPT-4, Claude 3.5/3.7) across diverse edge hardware configurations, we identify actionable reliability thresholds for production deployment. Our analysis reveals that procedural reliability, particularly tool initialization failures, constitutes the primary bottleneck for smaller models, while qwen2.5:32b achieves flawless performance matching GPT-4.1. The framework demonstrates that mid-sized models (qwen2.5:14b) offer practical accuracy-efficiency trade-offs on commodity hardware (96.6\% success rate, 7.3 s latency), enabling cost-effective intelligent agent deployment for resource-constrained organizations. This work establishes foundational infrastructure for systematic reliability evaluation of tool-augmented multi-agent AI systems.

Same-model self-verification, prompting a model to audit its own predicted answer, is a plausible confidence signal for selective prediction, but its practical value remains unclear once strong likelihood-based baselines are taken seriously. We evaluate self-verification against two such baselines, LL-AVG and LL-SUM, on ARC-Challenge and TruthfulQA-MC across multiple model families, scales, and prompt variants. We measure not only correctness ranking, but also abstention quality through AURC and operating-point analyses. The results are sharply task- and model-dependent. On ARC-Challenge, self-verification substantially improves over LL-AVG for Phi-2 and the Qwen models, with the largest gains appearing in Qwen-7B. On TruthfulQA-MC, however, the signal is less reliable: smaller models can become prompt-sensitive, DeepSeek-R1-Distill-8B degrades relative to LL-AVG, and LL-SUM often remains the stronger practical baseline. We therefore do not treat self-verification as a general-purpose uncertainty estimator. In this setting, it is better understood as a conditional confidence signal whose value depends on task type, model family, prompt formulation, and, crucially, the baseline it must beat.

Long chain-of-thought (CoT) traces are widely used as supervision for reasoning-oriented LLM SFT, yet answer-correct traces can still lead to markedly different fine-tuning outcomes. We study post-conclusion continuation in answer-correct long-CoT data: a continuation where the answer appears sufficiently supported, but the trace continues with additional reasoning that remains in the supervised target. To test its training effect, we use a delete-only editor to construct answer-preserving suffix removal and compare CoT-based SFT on the original and processed traces. We observe improved SFT outcomes after removing the editor-identified post-conclusion continuation, suggesting that this continuation is harmful to training in our setting. We therefore refer to this empirically supported phenomenon as harmful continuation. Beyond this intervention, we further characterize the removed post-conclusion continuation through uncertainty and hidden-state progress. We observe persistent local uncertainty together with weakened terminal-directional progress, forming an uncertainty--geometry mismatch. Finally, we instantiate Harmful Continuation Cut (HCC), a lightweight boundary proxy that approximates the editor-identified post-conclusion continuation boundary.

In order to minimize the generalization error in neural networks, a novel technique to identify overfitting phenomena when training the learner is formally introduced. This enables support of a reliable and trustworthy early stopping condition, thus improving the predictive power of that type of modeling. Our proposal exploits the correlation over time in a collection of online indicators, namely characteristic functions for indicating if a set of hypotheses are met, associated with a range of independent stopping conditions built from a canary judgment to evaluate the presence of overfitting. That way, we provide a formal basis for decision making in terms of interrupting the learning process. As opposed to previous approaches focused on a single criterion, we take advantage of subsidiarities between independent assessments, thus seeking both a wider operating range and greater diagnostic reliability. With a view to illustrating the effectiveness of the halting condition described, we choose to work in the sphere of natural language processing, an operational continuum increasingly based on machine learning. As a case study, we focus on parser generation, one of the most demanding and complex tasks in the domain. The selection of cross-validation as a canary function enables an actual comparison with the most representative early stopping conditions based on overfitting identification, pointing to a promising start toward an optimal bias and variance control.

Safe deployment of large language models (LLMs) may benefit from a reliable method for assessing their generated content to determine when to abstain or to selectively generate. While likelihood-based metrics such as perplexity are widely employed, recent research has demonstrated the limitations of using sequence-level probability estimates given by LLMs as reliable indicators of generation quality. Conversely, LLMs have demonstrated strong calibration at the token level, particularly when it comes to choosing correct answers in multiple-choice questions or evaluating true/false statements. In this work, we reformulate open-ended generation tasks into token-level prediction tasks, and leverage LLMs' superior calibration at the token level. We instruct an LLM to self-evaluate its answers, employing either a multi-way comparison or a point-wise evaluation approach, with the option to include a ``None of the above'' option to express the model's uncertainty explicitly. We benchmark a range of scoring methods based on self-evaluation and evaluate their performance in selective generation using TruthfulQA and TL;DR. Through experiments with PaLM-2 and GPT-3, we demonstrate that self-evaluation based scores not only improve accuracy, but also correlate better with the overall quality of generated content.

Large language models perform strongly on benchmarks in mathematical reasoning, coding and document analysis, suggesting a broad ability to follow instructions. However, it remains unclear whether such success reflects general logical competence, repeated application of learned procedures, or pattern matching that mimics rule execution. We investigate this question by introducing Stable Counting Capacity, an assay in which models count repeated symbols until failure. The assay removes knowledge dependencies, semantics and ambiguity from evaluation, avoids lexical and tokenization confounds, and provides a direct measure of procedural reliability beyond standard knowledge-based benchmarks. Here we show, across more than 100 model variants, that stable counting capacity remains far below advertised context limits. Model behavior is consistent neither with open-ended logic nor with stable application of a learned rule, but instead with use of a finite set of count-like internal states, analogous to counting on fingers. Once this resource is exhausted, the appearance of rule following disappears and exact execution collapses into guessing, even with additional test-time compute. These findings show that fluent performance in current language models does not guarantee general, reliable rule following.

In many social-science research tasks, such as economics, LLM-based agents must produce outputs for which no cheap, task-complete, machine-readable correctness signal exists. This creates a distinctive reliability problem for multi-agent systems: how should generation, critique, coordination, and human judgment be organized when no component can certify the final result? We address this problem through pAI-Econ-claude, a gated, human-in-the-loop multi-agent architecture for AI-assisted economic theory development. Agents coordinate through a shared workspace of inspectable intermediate records; specialized gates diagnose targeted failure modes and recommend loopbacks without certifying correctness; and human checkpoints retain authority over decisions that are costly to reverse. We evaluate the architecture on five matched economic-theory tasks against an ungated baseline. Two evaluators blinded to configuration agreed on all five pairwise rankings, preferring the gated architecture in four tasks and the baseline in one. Mean failure severity fell from 1.58 to 1.16, while overall usefulness rose from 2.60 to 3.10. The largest observed gain occurred when a reality check rejected a false market-structure premise and a proof review prompted revision of a false welfare claim. The negative case shows that scaffolding can also compress an economically important mechanism too aggressively. The results support a bounded claim: gated oversight improves the auditability of AI-assisted economic theory without substituting for formal verification, and the allocation of irreversible human judgment is a more informative design variable than pure agent autonomy. The workflow is publicly available at https://github.com/maxwell2732/pAI-Econ-claude.

Skills are becoming a reusable operational layer for LLM agents, encoding SOPs, domain rules, tool workflows, scripts, and validation routines. In realistic skill repositories, overlapping skills make reliable skill-use difficult. Final verifier success is too coarse for both evaluation and training, since an agent may pass through trial and error while selecting distractor skills, skipping required steps, composing workflows incorrectly or omitting final checks. We introduce SkillCoach, a self-evolving rubric framework for evaluating and enhancing agentic skill-use. SkillCoach derives skill-grounded process rubrics from real rollouts and evaluates trajectories along four dimensions: skill selection, skill following, skill composition, and skill-grounded reflection. It keeps the external verifier as a separate outcome signal, allowing process quality to be distinguished from accidental task success. The evolved rubrics further serve as process supervision for selecting high-quality training trajectories. Experiments show that evolved rubrics substantially improve evaluation quality, expose failures hidden by final accuracy, and provide stronger supervision signals than outcome-only filtering for enhancing agentic skill-use.

Clawdbot is a self-hosted, tool-using personal AI agent with a broad action space spanning local execution and web-mediated workflows, which raises heightened safety and security concerns under ambiguity and adversarial steering. We present a trajectory-centric evaluation of Clawdbot across six risk dimensions. Our test suite samples and lightly adapts scenarios from prior agent-safety benchmarks (including ATBench and LPS-Bench) and supplements them with hand-designed cases tailored to Clawdbot's tool surface. We log complete interaction trajectories (messages, actions, tool-call arguments/outputs) and assess safety using both an automated trajectory judge (AgentDoG-Qwen3-4B) and human review. Across 34 canonical cases, we find a non-uniform safety profile: performance is generally consistent on reliability-focused tasks, while most failures arise under underspecified intent, open-ended goals, or benign-seeming jailbreak prompts, where minor misinterpretations can escalate into higher-impact tool actions. We supplemented the overall results with representative case studies and summarized the commonalities of these cases, analyzing the security vulnerabilities and typical failure modes that Clawdbot is prone to trigger in practice.

Recent large-scale neural autoregressive sequence models have shown impressive performances on a variety of natural language generation tasks. However, their generated sequences often exhibit degenerate properties such as non-termination, undesirable repetition, and premature termination, when generated with decoding algorithms such as greedy search, beam search, top-k sampling, and nucleus sampling. In this paper, we focus on the problem of non-terminating sequences resulting from an incomplete decoding algorithm. We first define an incomplete probable decoding algorithm which includes greedy search, top-k sampling, and nucleus sampling, beyond the incomplete decoding algorithm originally put forward by Welleck et al. (2020). We then propose a non-monotonic self-terminating language model, which significantly relaxes the constraint of monotonically increasing termination probability in the originally proposed self-terminating language model by Welleck et al. (2020), to address the issue of non-terminating sequences when using incomplete probable decoding algorithms. We prove that our proposed model prevents non-terminating sequences when using not only incomplete probable decoding algorithms but also beam search. We empirically validate our model on sequence completion tasks with various architectures.

Large language models (LLMs) deployed for logical reasoning in knowledge-intensive domains exhibit a subtle but critical failure: coherence can be vacuously achieved through systematic abstention. A model that withholds commitment to either entailment or refutation satisfies negation consistency while providing no utility. We introduce Coherence Under Commitment (CUC), a dual-query evaluation paradigm that jointly measures consistency and decisiveness. CUC contributes three innovations: (1) a commitment score c(φ) = p(φ) + p(lnotφ) quantifying probability mass allocated to decisive outcomes; (2) a deterministic elicitation protocol via normalized YES/NO log probabilities, eliminating sampling variance; and (3) a 3-way decision framework (True/False/Uncertain) operationalizing the coherence-commitment trade-off into metrics. Experiments on four open-weight LLMs (1B-3B) across 204 FOLIO examples expose a sharp frontier. Qwen2.5-3B achieves near-zero contradiction (E[v_{neg}]{=}0.025) but only 7.4% coverage, while TinyLlama-1.1B reaches 79.4% coverage with violations on every example. Coherence-only evaluation would rank the abstaining model first; CUC exposes this as vacuous, and the frontier generalizes to LogiQA~v2 (ρ{=}0.97). We argue that evaluation must report both coherence and non-vacuous commitment and release a toolkit for standardized assessment.

Many deployments must compare candidate language models for safety before a labeled benchmark exists for the relevant language, sector, or regulatory regime. We formalize this setting as benchmarkless comparative safety scoring and specify the contract under which a scenario-based audit can be interpreted as deployment evidence. Scores are valid only under a fixed scenario pack, rubric, auditor, judge, sampling configuration, and rerun budget. Because no labels are available, we replace ground-truth agreement with an instrumental-validity chain: responsiveness to a controlled safe-versus-abliterated contrast, dominance of target-driven variance over auditor and judge artifacts, and stability across reruns. We instantiate the chain in SimpleAudit, a local-first scoring instrument, and validate it on a Norwegian safety pack. Safe and abliterated targets separate with AUROC values between 0.89 and 1.00, target identity is the dominant variance component (η^2 approx 0.52), and severity profiles stabilize by ten reruns. Applying the same chain to Petri shows that it admits both tools. The substantial differences arise upstream of the chain, in claim-contract enforcement and deployment fit. A Norwegian public-sector procurement case comparing Borealis and Gemma 3 demonstrates the resulting evidence in practice: the safer model depends on scenario category and risk measure. Consequently, scores, matched deltas, critical rates, uncertainty, and the auditor and judge used must be reported together rather than collapsed into a single ranking.

Large Language Models used in ChatGPT have traditionally been trained to learn a refusal boundary: depending on the user's intent, the model is taught to either fully comply or outright refuse. While this is a strong mitigation for explicitly malicious prompts, focusing safety training on refusals can lead to brittleness for prompts with obscured user intent. Binary refusal boundaries are especially ill-suited for dual-use cases (such as biology or cybersecurity), where a user request can be answered safely at a high level, but in some cases can lead to malicious uplift if sufficiently detailed or actionable. As an alternative, we propose safe-completions: a safety-training approach that centers on the safety of the assistant's output, rather than a binary classification of the user's intent. Safe-completions seek to maximize helpfulness within the safety policy's constraints. We incorporated this approach into GPT-5 and find that across both production comparisons and internally controlled experiments, safe-completion training improves safety (especially on dual-use prompts), reduces the severity of residual safety failures, and substantially increases model helpfulness.

Ensuring the reliability of agent architectures and effectively identifying problematic agents when failures occur are crucial challenges in multi-agent systems (MAS). Advances in large language models (LLMs) have established LLM-based agents as a major branch of MAS, enabling major breakthroughs in complex problem solving and world modeling. However, the reliability implications of this shift remain largely unexplored. i.e., whether substituting traditional agents with LLM-based agents can effectively enhance the reliability of MAS. In this work, we investigate and quantify the reliability of LLM-based agents from the perspective of Byzantine fault tolerance. We observe that LLM-based agents demonstrate stronger skepticism when processing erroneous message flows, a characteristic that enables them to outperform traditional agents across different topological structures. Motivated by the results of the pilot experiment, we design CP-WBFT, a confidence probe-based weighted Byzantine Fault Tolerant consensus mechanism to enhance the stability of MAS with different topologies. It capitalizes on the intrinsic reflective and discriminative capabilities of LLMs by employing a probe-based, weighted information flow transmission method to improve the reliability of LLM-based agents. Extensive experiments demonstrate that CP-WBFT achieves superior performance across diverse network topologies under extreme Byzantine conditions (85.7\% fault rate). Notably, our approach surpasses traditional methods by attaining remarkable accuracy on various topologies and maintaining strong reliability in both mathematical reasoning and safety assessment tasks.

Run the same LLM agent on the same task twice: do you get the same behavior? We find the answer is often no. In a study of 3,000 agent runs across three models (Llama 3.1 70B, GPT-4o, and Claude Sonnet 4.5) on HotpotQA, we observe that ReAct-style agents produce 2.0--4.2 distinct action sequences per 10 runs on average, even with identical inputs. More importantly, this variance predicts failure: tasks with consistent behavior (leq2 unique paths) achieve 80--92% accuracy, while highly inconsistent tasks (geq6 unique paths) achieve only 25--60%, a 32--55 percentage point gap depending on model. We trace variance to early decisions: 69% of divergence occurs at step 2, the first search query. Our results suggest that monitoring behavioral consistency during execution could enable early error detection and improve agent reliability.

Agentic systems are evaluated on benchmarks where agents interact with environments to solve tasks. Most papers report a pass@1 score computed from a single run per task, assuming this gives a reliable performance estimate. We test this assumption by collecting 60,000 agentic trajectories on SWE-Bench-Verified, spanning three models and two scaffolds. We find substantial variance: single-run pass@1 estimates vary by 2.2 to 6.0 percentage points depending on which run is selected, with standard deviations exceeding 1.5 percentage points even at temperature 0. This variance has critical implications: reported improvements of 2--3 percentage points may reflect evaluation noise rather than genuine algorithmic progress. Through token-level analysis, we show that trajectories diverge early, often within the first few percent of tokens, and that these small differences cascade into different solution strategies. To enable reliable evaluation of agentic systems, we recommend three concrete practices: (1) estimate pass@1 from multiple independent runs per task, especially when measuring small improvements, (2) use statistical power analysis to determine the number of runs needed to detect expected effect sizes, and (3) consider metrics like pass@k (optimistic bound) and pass^k (pessimistic bound) with k>1 to better characterize the full performance envelope. While these practices increase evaluation cost, they are essential for distinguishing genuine scientific progress from statistical noise.

As modern microservice systems grow increasingly popular and complex-often consisting of hundreds or even thousands of fine-grained, interdependent components-they are becoming more susceptible to frequent and subtle failures. Ensuring system reliability therefore hinges on accurate and efficient failure localization. Traditional failure localization approaches based on small models lack the flexibility to adapt to diverse failure scenarios, while recent LLM-based methods suffer from two major limitations: they often rely on rigid invocation workflows that constrain the model's ability to dynamically explore optimal localization paths, and they require resource-intensive inference, making them cost-prohibitive for real-world deployment. To address these challenges, we explore the use of reinforcement fine-tuning to equip lightweight LLMs with reasoning and self-refinement capabilities, significantly improving the cost-effectiveness and adaptability of LLM-based failure localization. We begin with an empirical study to identify three key capabilities essential for accurate localization. Building on these insights, we propose a progressive multi-stage GRPO fine-tuning framework, which integrates a multi-factor failure localization grader and a recursion-of-thought actor module. The resulting model, ThinkFL, not only outperforms existing state-of-the-art LLMs and baseline methods in localization accuracy but also reduces end-to-end localization latency from minutes to seconds, demonstrating strong potential for real-world applications.

Interactive agent benchmarks face a tension between scalable construction and realistic workflow evaluation. Hand-authored tasks are expensive to extend and revise, while static prompt evaluation misses failures that only appear when agents operate over persistent state. Existing interactive benchmarks have advanced agent evaluation significantly, but most initialize tasks from clean state and do not systematically test how agents handle pre-existing partial, stale, or conflicting artifacts. We present ClawForge, a generator-backed benchmark framework for executable command-line workflows under state conflict. The framework compiles scenario templates, grounded slots, initialized state, reference trajectories, and validators into reproducible task specifications, and evaluates agents step by step over persistent workflow surfaces using normalized end state and observable side effects rather than exact trajectory matching. We instantiate this framework as the ClawForge-Bench (17 scenarios, 6 ability categories). Results across seven frontier models show that the best model reaches only 45.3% strict accuracy, wrong-state replacement remains below 17\% for all models, and the widest model separation (17% to 90%) is driven by whether agents inspect existing state before acting. Partial-credit and step-efficiency analyses further reveal that many failures are near-miss closures rather than early breakdowns, and that models exhibit qualitatively different failure styles under state conflict.

A reliable executable environment is the foundation for ensuring that large language models solve software engineering tasks. Due to the complex and tedious construction process, large-scale configuration is relatively inefficient. However, most methods always overlook fine-grained analysis of the actions performed by the agent, making it difficult to handle complex errors and resulting in configuration failures. To address this bottleneck, we propose EvoConfig, an efficient environment configuration framework that optimizes multi-agent collaboration to build correct runtime environments. EvoConfig features an expert diagnosis module for fine-grained post-execution analysis, and a self-evolving mechanism that lets expert agents self-feedback and dynamically adjust error-fixing priorities in real time. Empirically, EvoConfig matches the previous state-of-the-art Repo2Run on Repo2Run's 420 repositories, while delivering clear gains on harder cases: on the more challenging Envbench, EvoConfig achieves a 78.1% success rate, outperforming Repo2Run by 7.1%. Beyond end-to-end success, EvoConfig also demonstrates stronger debugging competence, achieving higher accuracy in error identification and producing more effective repair recommendations than existing methods.

AI agents are increasingly deployed to execute important tasks. While rising accuracy scores on standard benchmarks suggest rapid progress, many agents still continue to fail in practice. This discrepancy highlights a fundamental limitation of current evaluations: compressing agent behavior into a single success metric obscures critical operational flaws. Notably, it ignores whether agents behave consistently across runs, withstand perturbations, fail predictably, or have bounded error severity. Grounded in safety-critical engineering, we provide a holistic performance profile by proposing twelve concrete metrics that decompose agent reliability along four key dimensions: consistency, robustness, predictability, and safety. Evaluating 14 agentic models across two complementary benchmarks, we find that recent capability gains have only yielded small improvements in reliability. By exposing these persistent limitations, our metrics complement traditional evaluations while offering tools for reasoning about how agents perform, degrade, and fail.

Error detection (ED), which aims to identify incorrect or inconsistent cell values in tabular data, is important for ensuring data quality. Recent state-of-the-art ED methods leverage the pre-trained knowledge and semantic capability embedded in large language models (LLMs) to directly label whether a cell is erroneous. However, this LLM-as-a-labeler pipeline (1) relies on the black box, implicit decision process, thus failing to provide explainability for the detection results, and (2) is highly sensitive to prompts, yielding inconsistent outputs due to inherent model stochasticity, therefore lacking robustness. To address these limitations, we propose an LLM-as-an-inducer framework that adopts LLM to induce the decision tree for ED (termed TreeED) and further ensembles multiple such trees for consensus detection (termed ForestED), thereby improving explainability and robustness. Specifically, based on prompts derived from data context, decision tree specifications and output requirements, TreeED queries the LLM to induce the decision tree skeleton, whose root-to-leaf decision paths specify the stepwise procedure for evaluating a given sample. Each tree contains three types of nodes: (1) rule nodes that perform simple validation checks (e.g., format or range), (2) Graph Neural Network (GNN) nodes that capture complex patterns (e.g., functional dependencies), and (3) leaf nodes that output the final decision types (error or clean). Furthermore, ForestED employs uncertainty-based sampling to obtain multiple row subsets, constructing a decision tree for each subset using TreeED. It then leverages an Expectation-Maximization-based algorithm that jointly estimates tree reliability and optimizes the consensus ED prediction. Extensive xperiments demonstrate that our methods are accurate, explainable and robust, achieving an average F1-score improvement of 16.1% over the best baseline.

Failures in multi-turn reasoning models are largely invisible to terminal-score evaluation. A model can lock onto an unsafe stance early in a long dialogue, yet its final-turn refusal rate may appear indistinguishable from a robustly aligned baseline. To expose these hidden temporal dynamics, we propose a trace-level diagnostic - the CoT-Output 2x2 safety matrix. This framework labels every turn along two independent axes (internal reasoning and visible output), yielding four operationally defined failure cells: robust alignment, alignment faking, overt jailbreak, and a distinct failure mode we term context-injection failure (where the CoT maintains safe reasoning, but the visible output produces harm, highlighting a multi-turn manifestation of reasoning unfaithfulness). We evaluate three distilled reasoning targets against a fixed attacker across five oversight conditions, collecting 6750 turn-level observations on the Information-Hazard scenario. Our analysis reveals two reproducible vulnerabilities: an oversight paradox where explicit monitoring cues paradoxically increase alignment-faking rates rather than suppress them, and a context-injection failure where models lock onto unsafe external outputs despite safe internal states. We release the full dataset of multi-turn dialogues and CoT traces to support follow-up trace-diagnostic research.

The continuous delivery of modern software requires the execution of many automated pipeline jobs. These jobs ensure the frequent release of new software versions while detecting code problems at an early stage. For TELUS, our industrial partner in the telecommunications field, reliable job execution is crucial to minimize wasted time and streamline Continuous Deployment (CD). In this context, flaky job failures are one of the main issues hindering CD. Prior studies proposed techniques based on machine learning to automate the detection of flaky jobs. While valuable, these solutions are insufficient to address the waste associated with the diagnosis of flaky failures, which remain largely unexplored due to the wide range of underlying causes. This study examines 4,511 flaky job failures at TELUS to identify the different categories of flaky failures that we prioritize based on Recency, Frequency, and Monetary (RFM) measures. We identified 46 flaky failure categories that we analyzed using clustering and RFM measures to determine 14 priority categories for future automated diagnosis and repair research. Our findings also provide valuable insights into the evolution and impact of these categories. The identification and prioritization of flaky failure categories using RFM analysis introduce a novel approach that can be used in other contexts.

Mathematical reasoning models are widely deployed in education, automated tutoring, and decision support systems despite exhibiting fundamental computational instabilities. We demonstrate that state-of-the-art models (Qwen2.5-Math-7B) achieve 61% accuracy through a mixture of reliable and unreliable reasoning pathways: 18.4% of correct predictions employ stable, faithful reasoning while 81.6% emerge through computationally inconsistent pathways. Additionally, 8.8% of all predictions are silent failures -- confident yet incorrect outputs. Through comprehensive analysis using novel faithfulness metrics, we reveal: (1) reasoning quality shows weak negative correlation with correctness (r=-0.21, p=0.002), reflecting a binary classification threshold artifact rather than a monotonic inverse relationship; (2) scaling from 1.5B to 7B parameters (4.7x increase) provides zero accuracy benefit on our evaluated subset (6% of GSM8K), requiring validation on the complete benchmark; and (3) latent reasoning employs diverse computational strategies, with ~20% sharing CoT-like patterns. These findings highlight that benchmark accuracy can mask computational unreliability, demanding evaluation reforms measuring stability beyond single-sample metrics.

We propose a general two-stage algorithm that enjoys a provable scaling law for the test-time compute of large language models (LLMs). Given an input problem, the proposed algorithm first generates N candidate solutions, and then chooses the best one via a multiple-round knockout tournament where each pair of candidates are compared for K times and only the winners move on to the next round. In a minimalistic implementation, both stages can be executed with a black-box LLM alone and nothing else (e.g., no external verifier or reward model), and a total of N times (K + 1) highly parallelizable LLM calls are needed for solving an input problem. Assuming that a generated candidate solution is correct with probability p_{gen} > 0 and a comparison between a pair of correct and incorrect solutions identifies the right winner with probability p_{comp} > 0.5 (i.e., better than a random guess), we prove theoretically that the failure probability of the proposed algorithm decays to zero exponentially with respect to N and K: $P(final output is incorrect) le (1 - p_{gen})^N + lceil log_2 N rceil e^{-2 K (p_{comp} - 0.5)^2}.$ Our empirical results with the challenging MMLU-Pro benchmark validate the technical assumptions, as well as the efficacy of the proposed algorithm and the gains from scaling up its test-time compute.

Coding agents often finish a session by returning one squashed patch that mixes feature implementation, bug fixes, refactorings, tests, and configuration edits. While the final code may be correct, collapsing unrelated edits into one patch removes the history structure needed for review, selective revert, and later maintenance. We study retrospective commit-history reconstruction: given a completed squashed change, an agent groups its hunks into commits and materializes a replayable commit sequence. We formalize the task as hunk-to-commit partitioning with a replay requirement and build AtomicCommitBench, containing 800 real consecutive-commit episodes from 10 Python projects. Because multiple decompositions may be reasonable, we evaluate outputs using complementary metrics: PPAR for replay validity, ARI for reference-based grouping quality, and TCR for failure containment on scoreable modified-test episodes. Natural retrospective reconstruction proves substantially harder than replay checking or synthetic tangling. Although nearly all non-random methods achieve replay validity (PPAR >= 0.988), grouping quality ranges from 0.03 to 0.46 ARI. Matched synthetic composites are much easier than real same-author squashed diffs (+0.333 ARI). In our evaluation, the GPT-5.4 setup driven by Codex CLI (0.46 ARI) and the GLM-5 setup driven by Claude Code (0.43 ARI) outperform MiniMax (0.31) and Kimi (0.29). Qualitative analysis identifies same-file lumping and support-hunk drift as recurring failure modes. Dependency-Aware Commit Evidence (DACE) improves the lower-scoring setups by 0.05 to 0.08 ARI, indicating that dependency cues and hunk-role information help agents avoid locality-driven grouping errors. AtomicCommitBench enables evaluation of the commit histories produced by coding agents alongside the final code.

In industrial procurement, an LLM answer is useful only if it survives a standards check: recommended material must match operating condition, every parameter must respect a regulated threshold, and no procedure may contradict a safety clause. Partial correctness can mask safety-critical contradictions that aggregate LLM benchmarks rarely capture. We introduce IndustryBench, a 2,049-item benchmark for industrial procurement QA in Chinese, grounded in Chinese national standards (GB/T) and structured industrial product records, organized by seven capability dimensions, ten industry categories, and panel-derived difficulty tiers, with item-aligned English, Russian, and Vietnamese renderings. Our construction pipeline rejects 70.3% of LLM-generated candidates at a search-based external-verification stage, calibrating how unreliable industrial QA remains after LLM-only filtering.Our evaluation decouples raw correctness, scored by a Qwen3-Max judge validated at κ_w = 0.798 against a domain expert, from a separate safety-violation (SV) check against source texts. Across 17 models in Chinese and an 8-model intersection over four languages, we find: (i) the best system reaches only 2.083 on the 0--3 rubric, leaving substantial headroom; (ii) Standards & Terminology is the most persistent capability weakness and survives item-aligned translation; (iii) extended reasoning lowers safety-adjusted scores for 12 of 13 models, primarily by introducing unsupported safety-critical details into longer final answers; and (iv) safety-violation rates reshuffle the leaderboard -- GPT-5.4 climbs from rank 6 to rank 3 after SV adjustment, while Kimi-k2.5-1T-A32B drops seven positions.Industrial LLM evaluation therefore requires source-grounded, safety-aware diagnosis rather than aggregate accuracy. We release IndustryBench with all prompts, scoring scripts, and dataset documentation.

We introduce CausaLab, a scalable environment for evaluating interactive causal discovery by LLM agents. Unlike prior evaluations, CausaLab evaluates both whether an agent can solve a problem using causal evidence and whether its answer is grounded in a faithful recovered causal mechanism. Each episode places an agent in a synthetic laboratory: it receives prior measurement records, intervenes on a manipulator crystal, and predicts the resonance frequency of a held-out reactor crystal governed by the same mechanism. The hidden data-generating process is a randomly sampled structural causal model (SCM), so success requires recovering both a causal graph and structural equations rather than recalling prior knowledge. Experiments show a persistent gap between prediction and mechanism recovery: in the purely observational 6-node setting, GPT-5.2-high reaches 92% task accuracy but only 0.471 all-edge F_1. Mixed observation-intervention strategies improve structural fidelity, while pure intervention remains difficult even for strong agents. We identify premature stopping as a major weakness and show that consistency verification mitigates it. CausaLab therefore separates predictive success from causal understanding and exposes current LLM agents' limits as experimental causal reasoners.

Language models are increasingly capable and are being rapidly deployed on a population-level scale. As a result, the safety of these models is increasingly high-stakes. Fortunately, advances in alignment have significantly reduced the likelihood of harmful model outputs. However, when models are queried billions of times in a day, even rare worst-case behaviors will occur. Current safety evaluations focus on capturing the distribution of inputs that yield harmful outputs. These evaluations disregard the probabilistic nature of models and their tail output behavior. To measure this tail risk, we propose a method to efficiently estimate the probability of harmful outputs for any input query. Instead of naive brute-force sampling from the target model, where harmful outputs could be rare, we operationalize importance sampling by creating unsafe versions of the target model. These unsafe versions enable sample-efficient estimation by making harmful outputs more probable. On benchmarks measuring misuse and misalignment, these estimates match brute-force Monte Carlo estimates using 10-20x fewer samples. For example, we can estimate probability of harmful outputs on the order of 10^-4 with just 500 samples. Additionally, we find that these harmfulness estimates can reveal the sensitivity of models to perturbations in model input and predict deployment risks. Our work demonstrates that accurate rare-event estimation is both critical and feasible for safety evaluations. Code is available at https://github.com/rangell/LMTailRisk

As LLM-based Multi-Agent Systems (MAS) are increasingly deployed for complex tasks, ensuring their reliability has become a pressing challenge. Since MAS coordinate through unstructured natural language rather than rigid protocols, they are prone to semantic failures (e.g., hallucinations, misinterpreted instructions, and reasoning drift) that propagate silently without raising runtime exceptions. Prevailing evaluation approaches, which measure only end-to-end task success, offer limited insight into how these failures arise or how effectively agents recover from them. To bridge this gap, we propose MAS-FIRE, a systematic framework for fault injection and reliability evaluation of MAS. We define a taxonomy of 15 fault types covering intra-agent cognitive errors and inter-agent coordination failures, and inject them via three non-invasive mechanisms: prompt modification, response rewriting, and message routing manipulation. Applying MAS-FIRE to three representative MAS architectures, we uncover a rich set of fault-tolerant behaviors that we organize into four tiers: mechanism, rule, prompt, and reasoning. This tiered view enables fine-grained diagnosis of where and why systems succeed or fail. Our findings reveal that stronger foundation models do not uniformly improve robustness. We further show that architectural topology plays an equally decisive role, with iterative, closed-loop designs neutralizing over 40% of faults that cause catastrophic collapse in linear workflows. MAS-FIRE provides the process-level observability and actionable guidance needed to systematically improve multi-agent systems.

In physics and engineering literature, the distribution of the excursion-above-zero time distribution (exceedance distribution) for a stationary Gaussian process has been approximated by a stationary switching process with independently distributed switching times. The approach matched the covariance of the clipped Gaussian process with the one for the stationary switching process and the distribution of the latter was used as the so-called independent interval approximation (IIA). The approach successfully assessed the persistency exponent for many physically important processes but left an unanswered question when such an approach leads to a mathematically meaningful and proper exceedance distribution. Here we address this question by proposing an alternative matching of the expected values of the clipped Slepian process and the corresponding switched process initiated at the origin. The method has allowed resolving the mathematical correctness of the matching method for a large subclass of the Gaussian processes with monotonic covariance, for which we provide a sufficient condition for the validity of the IIA. Within this class, the IIA produces a valid distribution for the excursion time and is represented in an explicit stochastic form that connects directly to the covariance of the underlying Gaussian process. We compare the excursion level distributions as well as the corresponding persistency exponents obtained through the IIA method with numerically computed exact distributions, and the simulated distribution for several important Gaussian models. We also argue that for stationary Gaussian processes with a non-monotonic covariance, the IIA fails and should not be used.

Combined electric power system and High-Altitude Electromagnetic Pulse (HEMP) models are being developed to determine the effect of a HEMP on the US power grid. The work relies primarily on deterministic methods; however, it is computationally untenable to evaluate the E1 HEMP response of large numbers of grid components distributed across a large interconnection. Further, the deterministic assessment of these components' failures are largely unachievable. E1 HEMP laboratory testing of the components is accomplished, but is expensive, leaving few data points to construct failure models of grid components exposed to E1 HEMP. The use of Bayesian priors, developed using the subject matter expertise, combined with the minimal test data in a Bayesian inference process, provides the basis for the development of more robust and cost-effective statistical component failure models. These can be used with minimal computational burden in a simulation environment such as sampling of Cumulative Distribution Functions (CDFs).

We introduce ProofGrid, a benchmark suite for evaluating LLM reasoning through machine-checkable proofs rather than final answers alone. ProofGrid contains 15 tasks spanning proof writing, proof checking, proof masking, and proof gap-filling. Tasks are expressed in minimal formal notation, especially NDL, a compact natural-deduction language that fits in short prompts and supports precise, auditable verification. This yields mechanical, reproducible, and fine-grained evaluation rather than judgments by humans or LLMs. ProofGrid covers a calibrated difficulty spectrum, from foundational reasoning tests to structurally rich challenge tasks that no current model solves, while minimizing reliance on domain knowledge, solver delegation, and long-context artifacts. We also develop a comparative framework for reasoning benchmarks and use it to situate ProofGrid relative to existing work in terms of representation, verification guarantees, and reasoning depth. Methodologically, we introduce an instrumented proof-checking pipeline that tolerates minor surface deviations while locating the first substantive reasoning failure, improving measurement resolution and separating proof planning from low-level execution noise. Using this pipeline, we evaluate a broad range of open and proprietary models. Results show rapid progress but substantial remaining limits: frontier models perform well on several foundational tasks, yet difficult tasks, especially those requiring global combinatorial reasoning or low-level proof synthesis, remain far from solved. We also identify epistemic instability, where models generate flawed proofs yet correctly reject those local inferences in isolation, and formalize this with an Epistemic Stability Index. Finally, we complement accuracy with 2PL IRT analyses, Wright maps, and a normalized task-discrimination measure based on Fisher information.

When an LLM agent reads a confidential file, then writes a summary, then emails it externally, no single step is unsafe, but the sequence is a data leak. We call this safety drift: individually safe actions compounding into violations. Prior work has measured this problem; we predict it. SafetyDrift models agent safety trajectories as absorbing Markov chains, computing the probability that a trajectory will reach a violation within a given number of steps via closed form absorption analysis. A consequence of the monotonic state design is that every agent will eventually violate safety if left unsupervised (absorption probability 1.0 from all states), making the practical question not if but when, and motivating our focus on finite horizon prediction. Across 357 traces spanning 40 realistic tasks in four categories, we discover that "points of no return" are sharply task dependent: in communication tasks, agents that reach even a mild risk state have an 85% chance of violating safety within five steps, while in technical tasks the probability stays below 5% from any state. A lightweight monitor built on these models detects 94.7% of violations with 3.7 steps of advance warning at negligible computational cost, outperforming both keyword matching (44.7% detection, 55.9% false positive rate) and per step LLM judges (52.6% detection, 38.2% false positive rate) while running over 60,000x faster.

Rubric-based evaluation is widely used in LLM benchmarks and training pipelines for open-ended, less verifiable tasks. While prior work has demonstrated the effectiveness of rubrics using downstream signals such as reinforcement learning outcomes, there remains no principled way to diagnose how a rubric itself fails from such aggregated or downstream signals alone. To address this gap, we introduce RIFT: RubrIc Failure mode Taxonomy, a taxonomy for systematically characterizing failure modes in rubric composition and design. RIFT consists of eight failure modes organized into three high-level categories: Reliability Failures, Content Validity Failures, and Consequential Validity Failures. RIFT is developed using grounded theory by iteratively annotating rubrics drawn from five diverse data sources spanning general instruction following, code generation, creative writing, and expert-level deep research, until no new failure modes are identified. We evaluate the consistency of the taxonomy by measuring agreement among independent human annotators, observing fair agreement overall (87% pairwise agreement and 0.64 average Cohen's kappa). Finally, to support scalable diagnosis, we propose automated rubric quality metrics and show that they align with human failure-mode annotations, achieving up to 0.925 F1.