LLM Agent Reliability

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

90 papers in the last four weeks, up 109% on the four weeks before. 0.9% of all new papers.

Jul 13Week of Sep 28

Latest papers 558

Jun 11, 2026cs.SE

Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces

Reliable operation of multi-agent large language model (LLM) systems depends on debugging long execution traces, where the few causally decisive events are buried in unstructured logs of messages, routes, memory writes, and tool calls. The standard tool is counterfactual replay (rewind, edit, and re-run the trajectory to measure each event's effect), but its cost grows linearly with the number of candidate events, making exhaustive replay infeasible at scale. We frame trace debugging as a knowledge-based decision-support problem. Each trace is compiled into a structured event knowledge graph over routing, memory, tool-use, uncertainty, and latent evidence, and a calibrated predictor decides where a scarce replay budget should be spent. We do not propose a new replay oracle; we propose a method to predict its results without paying the replay cost. We formulate zero-replay counterfactual-effect prediction: given a trace under a fixed budget, predict which events the oracle would mark high-effect before any replay is performed. BranchPoint-Latent is a lightweight predictor over observable, structural, uncertainty, and latent features of the knowledge graph. Calibrated against a deterministic replay oracle across 37 trace families, a single learning-to-rank gradient-boosted predictor raises per-trace localization (Branch Recall@5) from 0.73 to 0.93 on held-out families at zero oracle-replay cost. Rather than claiming universal dominance, we characterize when cheap graph centrality suffices and when learned evidence is necessary. The result is an auditable, cost-efficient decision-support system for AI-reliability debugging, positioned explicitly on the cost-accuracy frontier with reproducible artifacts.

Dong Ho Kang, Hyeonjeong Cha, Daein Weon

aUstechlab, Republic of Korea · bEwha Womans University, Seoul, Republic of Korea · cKookmin University, Seoul, Republic of Korea

Multi-Agent LLM Systems Counterfactual Prediction LLM Agents LLM Agent Reliability

Jun 11, 2026cs.AI

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

LLM-based agents have shown increasing potential in automating scientific discovery. Given an optimizable metric and an execution environment, they can propose, validate, and iterate scientific solutions, and have produced results that outperform human-designed approaches. As model capabilities continue to improve, we argue that the bottleneck for autonomous scientific discovery is shifting from prescribing agent workflows to designing agent environments: the resources, constraints, and interfaces that shape agent behavior. We frame this as environment engineering: building environments that amplify productive behaviors, such as open-ended exploration, systematic artifact management, and inter-agent collaboration, while suppressing harmful behaviors, such as reward hacking and high-friction human oversight. We present EurekAgent, an environment-engineered agent system for metric-driven autonomous scientific discovery. EurekAgent engineers the environment along four dimensions: permissions engineering for bounded agent execution and isolated evaluation; artifact engineering for filesystem and Git-based collaboration; budget engineering for budget-aware exploration; and human-in-the-loop engineering for easy human supervision and intervention. EurekAgent sets new state-of-the-art results on multiple mathematics, kernel engineering, and machine learning tasks, including new state-of-the-art 26-circle packing results discovered with less than $11 in total API cost. We open-source our code and results, and call for environment engineering as a core research direction for developing reliable autonomous research agents.

Amy Xin, Jiening Siow, Junjie Wang +5

Department of Computer Science and Technology, Tsinghua University · School of Information, Renmin University of China

Autonomous Scientific Discovery Autonomous Research Agents LLM Agent Reliability

Jun 11, 2026cs.AI

Confidence Composition for Multiagent Language Model Systems

Multiagent language model systems, such as collaborative reasoning and debate, produce multiple correlated candidate answers and confidence signals. However, these signals are usually calibrated only at the individual agent level, and provide no principled confidence estimate for the system's final answer. We formulate this as a confidence composition problem where combining confidence across agents and reasoning stages while preserving both selective utility and probabilistic reliability. We study confidence-aware routing and log-odds pooling protocols that select among candidate answers and output a system-level confidence. Across five benchmarks, 30 heterogeneous and homogeneous model pairs, and two confidence estimators, our gated-fusion methods improve AUARC and reduce Brier score over single agent, standard debate, and selective debate baselines, while retaining competitive weighted F1-score as a correctness metric. We further show that our log-odds fusion is overconfident due to correlated intermediate signals. We propose a shared dependence discount that substantially improves reliability while preserving predictions.

Ali Elahi, Michael J. Curry, Barbara Di Eugenio

University of Illinois Chicago

Multi-Agent LLM Systems Probability Calibration Multi-Agent Debate LLM Agent Reliability

Jun 11, 2026cs.SE

Understanding the Rejection of Fixes Generated by Agentic Pull Requests -- Insights from the AIDev Dataset

AI coding agents are increasingly used to generate pull requests (PRs) that propose code fixes in software projects. From a first exploration of the AIDev dataset, we find that 46.41% of the fixes proposed by the agents Copilot, Devin, Cursor, and Claude are rejected. This represents a significant amount of wasted resources that require human reviews, verifications, and running tests and validations for fixes that are merely discarded. Our goal in this paper is to understand the failure modes of AI-agents, an understanding that is crucial for better integrating AI-agents as efficient teammates. In this paper, we conduct a qualitative study on a representative sample of 306 non-merged pull requests created or co-authored by the agents mentioned earlier, followed by a quantitative analysis of the reasons for rejection. Our qualitative findings identify 14 reasons divided into four high-level categories for rejecting AI-agent fixes. We observe that developers can reject fixes due to fixes whose implementation is incorrect (e.g., incomplete, wrong approach), fixes that do not pass the continuous integration (CI) pipelines and fail tests, fixes for which the agent is unable to perform the implementation (e.g., no code generated, sessions lost), and fixes whose priority is low. Our results shed light on the importance of better guiding the model at these levels: (1) proposing hints about the approach to follow for fixing an issue, (2) outlining constraints or limitations regarding the approaches that should not be taken, and (3) instructing the agent on how to validate the implementation through CI pipelines and without introducing a breaking change. Our results suggest the need for good prioritization of tasks so that generated fixes do not lead to wasted human review efforts or wasted agent resources (e.g., tokens, compute, or allowed number of requests).

Mahmoud Abujadallah, Ali Arabat, Mohammed Sayagh

École de Technologie Supérieure Montréal, Canada

Agentic Code Generation Human-AI Collaboration LLM Agent Reliability

Jun 11, 2026cs.AI

TwinBI: An Agentic Digital Twin for Efficient Augmented Interactions with Business Intelligence Dashboards

Business intelligence (BI) increasingly combines dashboard interaction with LLM-based assistance, but these two modes often fall out of sync during multi-step analysis. As users switch between direct dashboard manipulation and natural-language queries, it becomes difficult to preserve a consistent analytical state across filters, hierarchies, metrics, and chart context. We present TwinBI, an agentic digital-twin framework that couples an LLM-based agent system with an executable BI dashboard state. TwinBI unifies conversational interaction, dashboard manipulation, semantic grounding, and provenance tracking through a shared analytical state reconstructed from a unified interaction log. It also exposes artifacts such as schema views, SQL, logs, and an /insights command for state-grounded analytical summaries. We evaluate TwinBI in two complementary ways. In a controlled A/B benchmark with the same backbone agent, TwinBI improves exact-match accuracy from 43.3% to 63.3%, partial-credit accuracy from 48.3% to 70.8%, and substantially reduces timeout rate from 40.0% to 10.0% relative to Dashboard alone. In a usability study, participants benefited from the integrated dashboard-and-chat workflow, with high task accuracy, moderate workload, and favorable ratings for state-aware interaction mechanisms. These results suggest that TwinBI improves both agent-level analytical reliability and user-facing analytical support by turning visible dashboard state into richer actionable context. Our dataset and source code are available at: https://github.com/simonjisu/TwinBI

Jisoo Jang Wen-Syan Li

Graduate School of Data Science, Seoul National University

AI-Assisted Decision Making LLM Agents Digital Twins Human-AI Collaboration LLM Agent Reliability

Jun 11, 2026cs.AI

LLM-as-an-Investigator: Evidence-First Reasoning for Robust Interactive Problem Diagnosis

Large language models (LLMs) are increasingly used as interactive assistants for technical problem solving. However, when users provide incomplete descriptions or plausible but unverified explanations, LLMs may prematurely align with these assumptions and propose solutions before collecting sufficient evidence. We refer to this behavior as user-driven sycophancy: the tendency of an LLM to reinforce a user-provided hypothesis instead of testing alternative explanations. This paper introduces LLM-as-an-Investigator, an evidence-first agentic AI methodology for robust problem diagnosis. The approach is implemented through a Solution Investigator Agent, which estimates the ambiguity of an initial problem description, generates candidate hypotheses, asks targeted clarification questions, and updates hypothesis probabilities after each answer. Rather than producing an immediate response, the agent continues the investigation until the evidence makes one candidate explanation stronger than the alternatives. To evaluate the approach, we build a benchmark from solved technical forum threads in mechanical, electrical, and hydraulic domains. We use a three-agent evaluation pipeline in which a Problem-Solution Extractor Agent converts solved threads into structured cases, a Ground-Truth Evaluator Agent simulates the user while hiding the known solution, and the tested assistant attempts to recover the solution through dialogue. The experiments compare standard assistants, reasoning-oriented LLMs, and the proposed investigator-based model across LLM backbones. In addition to diagnostic accuracy, we analyze how standard assistants follow misleading user hypotheses in diagnostic cases. The results show that the proposed approach identifies the problem more accurately than direct prompting and reasoning-only baselines, while its evidence-first protocol helps reduce user-induced conversational bias.

Fabrizio Marozzo, Pietro Liò

University of Calabria, Italy · University of Cambridge, United Kingdom

LLM Sycophancy LLM Agent Evaluation LLM Agent Reliability

Jun 11, 2026cs.AI

(Human) Attention Is (Still) All You Need: Human oversight makes AI-assisted social science reliable

Large language models (LLMs) are increasingly used for tasks once reserved for trained researchers, including hypothesis generation, specification choice, and drafting conclusions. We argue that the reliability of AI-assisted research depends not only on model capability, but also on how cognitive labour is structured between humans and machines. We study this problem through Human-in-the-Loop Economic Research (HLER), a decision architecture based on pre-commitment, decision sequencing, accountability, and attention allocation. In a pre-specified 2*4 factorial experiment with 280 complete research runs across four datasets, an unconstrained multi-agent baseline produced critical failures in 72% of runs. Using the same underlying model, the same agent decomposition, and identical prompts for the shared reasoning agents, HLER reduced the failure rate to 16% by imposing three architectural commitments: LLMs reason but do not execute data work, data and estimation are handled deterministically, and three human decision gates bind the workflow. Fisher's exact test rejects equality of failure rates at p<0.001. Reliability gains were largest on the least publicly represented dataset, a Qing-dynasty population register, consistent with a task-based production model with Frechet-distributed output quality. An 80-run ablation suggests that deterministic computation and human gates contribute independently, with exploratory evidence of complementarity. We interpret HLER as a research harness rather than an autonomous AI scientist: it sharply reduces failures, makes residual weaknesses more visible, and prevents unreliable claims from being advanced as publication-ready outputs.

Chen Zhu, Xiaolu Wang, Weilong Zhang

China Agricultural University · University of Cambridge

AI Agent Reliability AI-Assisted Scientific Research Human-in-the-Loop AI LLM Reliability Human Oversight of AI Multi-Agent LLMs+1

Jun 11, 2026cs.PL

XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows

LLM-based multi-agent systems increasingly coordinate planning, reasoning, tool use, and human interaction, yet their reliability remains limited. A central source of this limitation is the underspecified prompt--harness boundary. Current systems lack a principled way to decide which workflow commitments should remain in prompts and which should become harness structure. We present \textbf{XFlow}, an executable protocol programming system for reliable multi-agent workflows, and \textbf{XPF} (XFlow Protocol Format), its domain-specific protocol programming language. XFlow occupies a middle position between prompt-only orchestration and markup-like workflow descriptions. XPF remains readable as a literate protocol, but it is compiled and executed as a program. Its design keeps informal semantic work inside actors while moving selected commitments into harness structure that can be checked, preserved, and enforced. At runtime, XFlow stages uncertainty through lifecycle-governed symbols, which are typed state cells with validation and commit states. Actor outputs are mediated before they become shared state, instead of spreading through prompts, transcripts, or implicit memory. Our experiments cover Constrained Interaction, Long-Context Reasoning, and Agentic Software Engineering. They show that XFlow improves reliability by making constraints, evidence handling, and process requirements explicit and enforceable.

Hanqi Li, Jing Peng, Zijian Wang +2

X-LANCE Lab, School of Computer Science, Shanghai Jiao Tong University, Shanghai, China · Suzhou Laboratory, Suzhou, China · Shanghai Innovation Institution, Shanghai, China +1

Multi-Agent Orchestration Agentic Workflows LLM Agent Reliability

Jun 10, 2026cs.AI

WorkBench Revisited: Workplace Agents Two Years On

The best agent on WorkBench in March 2024, GPT-4, completed just 43% of tasks. We revisit the benchmark in June 2026 and find that the best agent to date, Claude Fable 5, now completes 98%. Beyond this considerable progress in frontier agent performance, three things stand out. First, unintended harmful actions, such as emailing the wrong person, fell from 26% of tasks for GPT-4 to 1.9% for Claude Fable 5; capability and safety go together on WorkBench rather than trade off, so the models that finish the most tasks also do the least unintended damage. Second, the rise of open-weight models has drastically lowered costs for a performance level that was only accessible to proprietary models, while frontier costs have stayed stable. Third, while several classes of error have been eliminated, frontier models still make some basic mistakes that occasionally result in irreversible harm. We release an updated version of the benchmark with data and code quality improvements, new model scores, and analysis of agent progress on WorkBench since 2024.

Olly Styles, Sam Miller

Independent researcher

Computer-Use Agent Benchmarks AI Agent Reliability LLM Agent Evaluation AI Agent Benchmarks AI Agent Safety Benchmarks LLM Agent Safety+1

Jun 10, 2026cs.AI

Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents

Compact language models (LMs) reduce cost, latency, and deployment risk for tool agents. Yet MCP-style tool use requires more than isolated function calling: an agent must discover tools from live catalogs, satisfy schemas, preserve dependencies across intermediate outputs, and ground final responses in executed evidence. Small planners often generate plausible workflow graphs that fail under tool resolution, parameter validation, dependency tracking, or execution. We argue that this failure mode is poorly handled by small-corpus distillation. A few hundred teacher traces can teach workflow format, but rarely cover the recovery behavior needed to repair failed plans over changing tool catalogs. We introduce Evoflux, an inference-time evolutionary search method that treats compact tool use as the repair of executable tool workflows. It evolves typed workflow graphs through structured edits, execution feedback, adaptive intensity, meta-guided redesign, and diversity pruning. On held-out MCP-Bench tasks spanning live MCP servers and 250 tools, Evoflux raises execution feasibility from roughly 3% to 17-24% across small planners. In contrast, SFT and SFT+DPO on the same search-mined data match, underperform, or collapse below zero-shot performance; ReAct reaches higher peaks, but with higher variance and token cost. These results show that execution-grounded search is more reliable under scarce teacher-trace budgets.

Kushal Raj Bhandari, Ling Yue, Ching-Yun Ko +4

Rensselaer Polytechnic Institute, Troy, NY 12180 USA · IBM Research, Yorktown Heights, NY 10598 USA

Inference-Time Search LLM Agent Workflow Optimization Tool-Use Planning LLM Tool Use LLM Agent Reliability

Jun 10, 2026cs.CL

Notes2Skills: From Lab Notebooks to Certainty-Aware Scientific Agent Skills

Scientific discovery workflows rely heavily on lab notes, where researchers record observations, interpret uncertain results, and plan follow-up experiments. Unlike polished publications, lab notes preserve evolving scientific reasoning, tacit scientific knowledge, and author uncertainty, giving AI agents access to the process of science. However, most prior work on scientific text focuses on papers, protocols, or structured databases, leaving informal laboratory notes underexplored as inputs to AI agents for science. This gap matters because tacit knowledge in lab notes is not written as a clean set of instructions: validated observations, tentative judgments, and possible experimental next steps often appear in the same passage. If these signals are conflated, an AI agent may mistake uncertain scientific judgments for confirmed conclusions or executable actions. To this end, we present Notes2Skills, a two-stage framework for turning lab notebooks into verifiable skills for scientific AI agents while preserving the author's certainty. Across three wet-lab sessions and seven original downstream configurations, Notes2Skills is the only tested configuration that avoids both observed regime-level failures: over-acting on uncertain notes and missing firm directives. We show that certainty preservation is a key missing piece between lab notebooks and reliable agent skills, opening a path toward safer AI co-scientist systems.

Shi Liu, Jiayao Chen, Chengwei Qin +3

Southern University of Science and Technology, Shenzhen, China · Shenzhen Loop Area Institute, Shenzhen, China · The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China +1

AI for Science Document Information Extraction LLM Agent Skill Learning LLM Agent Reliability

Jun 10, 2026cs.CL

Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents

Long-horizon LLM agents are not trusted to run unattended: with no human watching, they confidently report success they never verified. We treat honesty -- bounding what an agent may claim at termination -- as a first-class metric for unattended autonomy, distinct from capability. We present Autopilot, an execution model that makes silent fabricated success structurally impossible rather than merely rarer. Autopilot externalizes all working state into a durable, gated finite-state machine that a scheduler advances one stateless tick at a time; a hard floor forbids any terminal "done" claim whose falsifiable gate did not actually execute and pass. We prove a No-False-Success theorem -- under gate soundness, floor enforcement, and plan coverage, termination implies the goal holds -- whose only trust points are empirically measurable, and show the worst case degrades to an honest stall, never a fabricated success. Because each tick rehydrates only the state machine, per-step context cost is constant in the horizon. Across a 3,150-cell paired corpus (70 tasks

\times

3 systems

\times

3 models

\times

5 seeds, including 50 SWE-bench Lite tasks across 11 OSS repos), Autopilot fabricates on 0.95% of cells [95% CI 0.38--1.62] while Reflexion and StateFlow baselines fabricate on 8.10% [6.48--9.81] and 25.05% [22.48--27.62] respectively. The headline contrast lives in the hard regime: on SWE-bench Lite, the firewall reduces fabrication from 33.7% (StateFlow) to 0.67%, a paired difference of

-33.07

pp [95% CI

-36.53, -29.73

]. The mechanism is the gate, not the model: all ten Autopilot fabrications come from the strongest model, while two weaker mid-tier models never fabricate across 700 paired cells. The firewall trades coverage for honesty by design -- an honest stall is recoverable; a confident wrong output shipped downstream is not.

Youwang Deng

EpistemicaLab — Independent Research

AI Agent Reliability LLM Hallucination Mitigation Long-Horizon Agent Evaluation LLM Agent Verification Runtime Enforcement for AI Agents Long-Horizon LLM Agents+1

Jun 10, 2026cs.AI

DFAH-Bench: Benchmarking Observable Agent Instability in Financial Decision-Making

A financial agent can repeat a decision while changing the work behind it. DFAH-Bench operationalizes the Determinism--Faithfulness Assurance Harness (DFAH), pairing decision agreement with tool-path agreement on the same qualified replays, then extends that qualification principle to evidence, authorization, execution and task outcomes. Retrospective and prospective replay analyses expose process variation behind stable decisions. Across 570 eligible prospective episodes, decision agreement is 94.2-95.1%, while agreement on ordered tools, arguments and results is 45.0-51.5%; one stratum falls one group below its prespecified coverage minimum. A separate capture diagnostic shows that systematic omissions can preserve perfect replay agreement. Using the

τ

-Knowledge banking environment, we retain 1,080 scheduled episodes and 1,033 known native outcomes across separate cohorts with open-weight and frontier generators. Missing outcomes prevented the planned tests, so comparisons are descriptive. On the primary schedule, structural checks alone yield more successes than either gate-and-recovery bundle. The typed-choice bundle has lower mean episode cost than the generative bundle on complete task pairs, but produces fewer successes under every assignment of unknown outcomes. Input limits and recovery behavior materially shape these results. Fixed-state probes reveal higher decision agreement alongside lower agreement with constructed policy labels, and separately expose sensitivity to retained generator rationale in a selected authorization case. Together, the findings connect replay observability to evidence, authorization, completion and cost: evidence sufficiency needs direct assessment alongside repeatability.

Raffi Khatchadourian

IBM Financial Services Market New York, NY, USA

Agent Evaluation AI Agent Benchmarks LLM Agent Reliability Financial Decision-Making

Jun 9, 2026cs.CL

Catching One in Five: LLM-as-Judge Blind Spots in Production Multi-Turn Transaction Agents

LLM-as-judge is the default instrument for evaluating conversational agents, yet its reliability is almost always reported as agreement with human ratings, not recall of real defects. We study a deployed multi-turn food-and-beverage ordering agent and measure how many genuine quality problems its built-in LLM judge catches, using exhaustive human transcript review as ground truth. Across three batches the judge surfaces well under a quarter of human-confirmed systematic problems -- 2 of 9 patterns (22%) in one batch, and its operational gate flagged zero of 100 rounds in a batch where humans confirmed 23 distinct defects and 7 new cross-cutting patterns. Our blind-spot taxonomy shows the failure is structured, not random: the judge catches turn-local issues (a fabricated statistic, a wrong language) but misses cross-turn state issues (confirm-gate lockout, cart hallucination, escalation lockout, stale referents). The mechanism: the scoring rubric exposes only three coarse axes (intent, brand-voice, personalization) and has no category for the behavioural dimensions -- state-tracking, guardrails, recovery -- where most defects cluster. The failure is routing, not perception: 113 of 114 rounds whose raw judge note describes a confirm-gate or cart-state defect are scored "brand voice", and none reach an operational failure -- the gate is wired to hangs and hard assertions, not the rubric -- so the 0% is a routing-and-wiring failure, not blindness. The consequence for prevalence estimation is sharp: when the apparent defect rate is zero the Rogan-Gladen correction degenerates -- no signal can recover the true rate -- while where the gate reports a nonzero rate the same estimator implies a 3-6x undercount under our measured sensitivity. For production multi-turn agents, automated judging is a regression floor, not a substitute for human review.

Sawyer Zhang, Alexander Wang, Sophie Lei

Lumivate (Lumi)

LLM-as-a-Judge LLM Agent Evaluation Human-in-the-Loop Evaluation LLM Agent Reliability

Jun 8, 2026cs.AI

Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents

Large language models deployed as autonomous agents for enterprise workflows face a key challenge: verbose tool responses from enterprise systems can cause context overflow, stale-state errors, and high inference cost. We study this problem in automated expense itemization in Microsoft Dynamics 365 Finance and Operations using Model Context Protocol tools. We evaluate four GPT-5 configurations on a 50-task hotel expense benchmark: no user model, full conversation history, context pruned to the last 5 tool call/response pairs, and pruning with automated summarization. Results are averaged across 5 independent runs, with the user model held constant for the context-engineering comparison. The no-user-model baseline achieves only 8.0% complete itemization. Full-context retention improves completion to 71.0%, but consumes 1,480,996 tokens and 14.56 hours per benchmark. Pruning to the last 5 tool calls improves completion to 79.0% while reducing token use to 535,274 and runtime to 5.39 hours. Adding summarization achieves the best result: 91.6% complete itemization and 99.64% average amount itemized, with 553,374 tokens and 5.79 hours. We further report confidence intervals, effect-size analysis, sensitivity over pruning and summary windows, failure analysis, results across five expense types grouped into three categories, and cross-model evidence with Claude Sonnet 4.5. These results show that, for this class of enterprise tool-use workflow, selective retention of recent tool interactions plus compact summarization can improve both reliability and efficiency compared with full-history retention.

Abhilasha Lodha, Mahsa Pahlavikhah Varnosfaderani, Abir Chakraborty +1

Microsoft

LLM Inference Efficiency Tool-Augmented Language Model Agents LLM Agent Workflow Optimization LLM Tool Use Long-Horizon LLM Agents LLM Context Management+2

Jun 8, 2026cs.IR

$τ$ -Rec: A Verifiable Benchmark for Agentic Recommender Systems

As recommender systems transition toward agentic, multi-turn conversational interfaces, evaluation paradigms have struggled to keep pace. Current benchmarks often rely on "LLM-as-a-judge" evaluations, which introduce subjectivity, high costs and inconsistency. We present

τ

-Rec, a benchmark for agentic recommender systems that replaces subjective evaluation with verifiable rewards and a reveal-tagged elicitation (RTE) mechanism that controls how task constraints surface during dialogue. By testing agents against structured catalog predicates and employing a pass^k reliability metric,

τ

-Rec provides a systematic test for consistent reasoning. Our evaluation of nine configurations across five model families -- GPT-5.4, Claude Sonnet 4.6, Gemini 2.5 Flash, DeepSeek V4 Flash, Qwen3-32B and GPT-5 mini -- reveals a steep reliability cliff, where even the best model achieves only ~57% at pass^1 and ~35% at pass^4, highlighting a critical gap in current conversational agent deployment. All code and data are publicly available at https://github.com/nbharaths/tau-rec.

Bharath Sivaram Narasimhan, Karthik R Narasimhan

Independent Researcher Mountain View, CA, USA · Princeton University Princeton, NJ, USA

AI Agent Reliability AI Agent Benchmarks Multi-Turn Dialogue Evaluation Recommender Systems LLM Agent Reliability

Jun 8, 2026cs.AI

Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters

Configuring an advanced scientific simulator, translating a modeling goal into a valid, runnable input deck, is a persistent bottleneck that costs domain scientists hours to days. Input decks are executable interfaces: simulator-specific vocabulary, cross-file references, schema constraints, and validation rules must align before a simulation can run. We show that this bottleneck can be substantially reduced with a lightweight adapter around an off-the-shelf coding agent, rather than a bespoke simulator agent. Coding agents already navigate files, edit code, run commands, and repair outputs; what they lack is the simulator's executable contract, and rebuilding the agent loop risks discarding harness-calibrated tool-use and self-correction behavior. We introduce SIGA, a coding-agent adapter that supplies this contract through retrieval, procedural memory, agent-callable validation, and validation-gated termination while leaving the model and loop frozen. Because this contract is small and external, SIGA also supports adapter self-evolution: prior trajectories can rewrite the adapter contents without modifying the underlying agent. On GEOS, a multiphysics subsurface simulator, SIGA's main gain is reliability: on harder held-out tasks it improves TreeSim from 0.720 to 0.789 and reduces across-run standard deviation by about 16x by preventing empty or invalid decks. In a human calibration, SIGA reaches in about five minutes the deck quality a domain expert reached in about three hours. Transfers to OpenFOAM and LAMMPS show the recipe is portable but interface-dependent: completion gates help when structural completeness is the bottleneck, while memory and retrieval help when value correctness is.

Matthew Ho, Brian Liu, Jixuan Chen +2

University of California, San Diego

Coding Agents Scientific Code Generation LLM Agent Reliability

Jun 8, 2026cs.AI

REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces

Large language model (LLM) agents now solve complex tasks through long plan-and-execution traces, yet the ability to locate errors in a completed traces still lags far behind, especially in the \emph{silent failure} regime. Existing approaches predict suspect steps via classifiers or LLM judges, or recover correct answers via retry, but none feed the intervention outcome back to \emph{refine the attribution itself}. We propose \methodname, a method that closes this gap by diagnosing a candidate error step, testing it through controlled replay with a diagnosis-specific patch, and using the verified outcome flip as contrastive evidence to refine the final attribution. Across four localization benchmarks spanning multi-hop reasoning across domains, \methodname achieves the highest localization accuracy among same-auditor methods across all four benchmarks, with the largest gains on structured tool-use traces, while providing actionable localization even when ground-truth answers are unavailable.

Xiaofeng Lin, Yingxu Wang, Tung Sum Thomas Kwok +4

University of California, Los Angeles, USA · Cisco, USA · University of Mississippi, USA

LLM Agent Evaluation LLM Agents Fault Localization LLM Agent Reliability

Jun 8, 2026cs.AI

The Token Not Taken: Sampling, State, and the Stochasticity of AI Agents

Agentic AI systems can behave differently across runs: the same request may produce a different plan, a different tool call, a different code edit, or a different final answer. Such variability arises from several layers that are often conflated. At the core of many current agents is a foundation model, a large pretrained model adaptable to many downstream tasks, embedded in an orchestration loop that plans, calls tools, observes results, and updates state. One explicit intrinsic source of variability in such systems is token generation: the model computes scores over possible next tokens, the scores are converted into probabilities, and a decoder may sample tokens using a pseudo-random number generator. A small sampled token difference can then cascade downstream into a different tool call, code path, search query, or agent state. Other sources of variability are extrinsic to token sampling, including changing environments, live data, serving infrastructure, batch effects, and numerical details. By separating these layers, this tutorial clarifies what it means to call agentic AI systems stochastic, when such variability can be reproduced under matched conditions, and why deterministic execution need not imply identical behavior in deployed settings.

Muhammad Zia Hydari, Raja Iqbal

University of Pittsburgh · Ejento AI Inc.

Language Model Decoding LLM Agents LLM Agent Reliability

Jun 6, 2026cs.LG

QueryWeaver: Reliable Multi-Tool Query Execution Planning via LLM-Based Graph Generation

Many real-world queries over personal data span multiple applications and require structured planning, as individual tools expose only partial information. While LLMs show strong reasoning and tool use, reliably executing multi-step, cross-tool queries remains challenging. We introduce a system that converts natural language queries into structured graphs and executes them via a deterministic planner. Our approach uses depth-first search to resolve dependencies and combine results across tools, improving reliability and enabling queries beyond traditional keyword-based search. We demonstrate high accuracy even with smaller or locally hosted LLMs.

Aishwarya Chakravarthy, Vidhi Kulkarni, Duen Horng Chau

School of Computational Science and Engineering, Georgia Institute of Technology, Atlanta, GA, USA.

Tool-Augmented Language Model Agents Tool-Use Planning LLM Tool Use Graph-Based Planning LLM Agent Reliability

Jun 6, 2026cs.MA

Silent Failure in LLM Agent Systems: The Entropy Principle and the Inevitable Disorder of Autonomous Agents

Large Language Model (LLM) agent systems suffer from failures that occur without external triggers -- no injection, no adversarial input, no resource exhaustion. These silent failures -- unexpected deviations from intended behavior under normal conditions -- are routinely misattributed to bugs or configuration errors. Through systematic analysis of over 40,000 controlled trials and long-term production observations spanning 100,000+ agent interactions, we identify a common structural logic underlying these failures. Building on patterns observed in our experiments, we survey the global research literature on autonomous agent reliability and synthesize 22 intrinsic properties of LLM agent systems across six lifecycle layers: foundation semantics, inter-agent transmission, memory persistence, task execution, feedback correction, and systemic evolution. We demonstrate that whenever a sufficient subset of these properties co-exist, system entropy -- the measurable accumulation of disorder: loss of output consistency, task accuracy, and cross-session coherence -- increases monotonically with interaction rounds. We formalize this as the Entropy Principle: S(t) = S0 * e^(alpha * t), with alpha measured empirically across multiple architectures. We propose the PIG (Physical Integrity Gate) Engine with the ADE (Agent Delivery Engineering) protocol suite as an engineering countermeasure to entropy-driven disorder. Our findings establish silent failure not as a bug to be fixed but as a manifestation of Intelligence Entropy -- a physical constraint to be managed through deterministic governance. We argue that any engineering effort stabilizing the structure and order of agent systems participates in a unified mission: keeping intelligent systems reliable as they grow in scale and complexity.

Dexing Liu

Shanghai Qijing Digital Technology Co., Ltd.

Agent Failure Analysis AI Agent Governance Long-Horizon LLM Agents LLM Agent Reliability

Jun 6, 2026cs.AI

PACE: Anytime-Valid Acceptance Tests for Self-Evolving Agents

Self-evolving agents improve by repeatedly proposing changes to their own prompts, skills, or workflows and keeping those that score higher on a small held-out set. Almost all effort has gone into the proposer that generates candidates; we argue the weak point is the acceptor, the rule that decides whether to commit a change. Applied hundreds of times against the same noisy dev estimate, the ubiquitous "keep it if the score went up" rule is uncontrolled adaptive multiple testing: the agent effectively p-hacks itself, accumulating false commits that make it churn and drift rather than improve. We recast committing as a sequential hypothesis test and propose PACE (Paired Anytime-valid Commit Evaluation), a training-free, anytime-valid commit gate. Each candidate is compared to the incumbent on identical instances and committed only when a testing-by-betting e-process accumulates decisive evidence, stopping early to save evaluations and controlling each candidate's false-commit probability at a user-set level even under optional stopping (a per-decision guarantee). On Qwen2.5 agents (0.5B-3B) self-evolving at the prompt level on GSM8K, SVAMP, and ARC-Challenge, greedy acceptance commits 30-42% false and 10-33% harmful edits when a genuine improvement is hidden among noisy proposals, while PACE commits the real one and essentially nothing else, matching greedy's held-out accuracy at sharply lower variance and about 18% lower evaluation cost. With no real gain available, greedy commits 13-21 spurious self-modifications per run (72-100% false) and degrades the most fragile agent by 4.9 points, while PACE holds at baseline. Reliability of self-evolution depends on the acceptor, not only on the proposer.

Zayx Shawn

Independent Researcher

Sequential Hypothesis Testing LLM Agent Reliability Multiple Hypothesis Testing Self-Evolving Agents

Jun 6, 2026cs.AI

SKILL.nb: Selective Formalization and Gated Execution for Durable Agent Workflows

AI agents increasingly turn past experience into reusable artifacts such as code, workflows, and procedural memories. Reuse can improve efficiency, but it also creates a lifecycle reliability problem: artifacts that succeed once may fail under environment drift, underspecified tasks, or changing task distributions, especially in web automation. We introduce SKILL.nb, a framework for governing reusable agent workflows with evidence-calibrated lifecycle policies. SKILL.nb uses selective formalization: execution evidence decides which workflow steps should become executable code, which should remain natural-language guided, and when those choices should be revised. Workflows are stored as auditable, versioned notebooks that interleave natural-language guidance, multi-language executable cells, validation gates, fallback paths, and multimodal evidence such as outputs, screenshots, and error traces. At runtime, gate-conditioned execution lets each step run code when its gates validate, or fall back locally when drift invalidates the executable realization. On WebArena-Verified, SKILL.nb achieves 53.7% single-round success, improving over the strongest baseline by 3.9 percentage points. Across three re-executions, it retains 91.7% of initially successful tasks, 15.5 points above the next best method. Under bounded repair, it recovers 72.9% of subsequent failures while limiting post-repair regressions to 4.2%, compared with 15.0% to 17.0% for persistent baselines. It also leads on Mind2Web cross-website and cross-domain splits. In a GitLab migration test, SKILL.nb preserves performance when reusing frozen state learned on GitLab 15.7, with frozen-versus-fresh target-version gaps of -1.7 points on GitLab 16.11 and +0.6 points on GitLab 18.9. These results identify lifecycle governance and gate-conditioned execution as reliability axes beyond one-shot task success.

Amine El Hattami, Nicolas Chapados, Christopher Pal

ServiceNow Research · Mila · Polytechnique Montréal +1

AI Agent Reliability Agentic Workflows Agent Skill Learning Web Agents LLM Agent Reliability

Jun 5, 2026cs.AI

Contract2Tool: Learning Preconditions and Effects for Reliable Tool-Augmented LLM Agents

Tool-augmented large language model agents increasingly rely on external APIs, but standard tool schemas describe how to call a tool, not when the tool is causally appropriate or what task state it produces. Causal tool filtering addresses this gap by using lightweight contracts that specify each tool's preconditions, effects, risk level, and cost. However, manually writing and maintaining such contracts does not scale to large or changing tool ecosystems. We introduce Contract2Tool, a framework for inferring tool contracts from metadata, schemas, documentation, and execution traces. Contract2Tool converts observable tool evidence into normalized symbolic contracts that can be evaluated intrinsically and deployed inside downstream causal tool filtering. We evaluate learned contracts against gold preconditions, effects, and risk labels, and measure their downstream utility on multi-step agent tasks. Our results show that hybrid documentation-and-trace evidence produces contracts accurate enough to preserve most of the reliability and efficiency benefits of gold contracts. Learned-contract CMTF achieves 0.980 downstream success, close to 0.990 for gold-contract CMTF, while reducing visible tools from 100 to 1 and reducing average token usage from 26,172 to 2,528 relative to all-tools exposure. These results suggest that learned contracts can provide a scalable contract layer between tool schemas and reliable agent execution.

Rahul Suresh Babu, Laxmipriya Ganesh Iyer

Independent Researcher United States of America

AI Agent Reliability Tool-Use Evaluation Tool-Using Agents LLM Agent Reliability

Jun 5, 2026cs.LG

Strained Coherence: A Pre-Failure Signal in Coding Agent Execution Trajectories

LLM-based coding agents sometimes acknowledge a problem in their own reasoning and then proceed anyway. We call this pattern strained coherence: a safety-relevant failure mode in which an agent has information that should change its behavior, states that information, and still acts against it. The pattern overlaps with verbalized reward hacking, where an agent names a tension between a task proxy and the underlying goal yet optimizes the proxy anyway. We give an operational definition, build a Claude Sonnet 4.6 judge that reads full trajectories and flags spans where the pattern occurs, and evaluate it on 44 Terminal-bench-2 trajectories using a Qwen3.5-35B-A3B backbone. Flagged trajectories fail 94% of the time versus 46% for unflagged trajectories (47-point gap, Fisher's exact p = 0.003; 46 points after excluding three prompt-embedded examples, p = 0.006). At matched selectivity, the detector reaches 94% precision versus 88% for a lexical discourse-marker baseline; the 10-trajectory intersection of the two methods has a 100% failure rate (Clopper-Pearson 95% CI [69%, 100%]). We replicate on Gemma4-31B with 43 trajectories: the overall signal is directionally consistent but not significant (20-point gap, p = 0.31), with attenuation driven largely by 13 trajectories with zero think content, where the detector has no substrate to analyze. In the high-verbosity Gemma tertile, the gap is +30 points; in the mid- and high-verbosity Qwen tertiles, it is +40 points each. The first flag appears at a median of 83-84% of elapsed trajectory time across both models, and the binary flag survives paraphrases that soften explicit conflict markers (8/8 trajectories). Unlike univariate predictors, the detector emits interpretable span-level output -- quoted acknowledgment, quoted action, and typed conflict -- showing what the agent saw and ignored.

Marut Pandya, Kasey Zhang, Baiqing Lyu

AI Coding Agents AI Agent Monitoring LLM Agent Reliability

Jun 5, 2026cs.AI

The Sim-to-Real Gap of Foundation Model Agents: A Unified MDP Perspective

Foundation model agents are increasingly deployed for real-world decision-making, but suffer from the sim-to-real gap. While robotics and classical control have mature frameworks to address this gap, the foundation model community is treating agent robustness as an entirely novel phenomenon. Our paper proposes formalizing the foundation model agent evaluation and training gap as a classical sim-to-real problem structured entirely around the four elements of a Markov Decision Process, including Observation, Action, Transition, and Reward. In this paper, we set a comprehensive research agenda that translates classical discrepancies into the foundation model domain and advocates for adopting established solutions like domain randomization. We provide concrete examples, such as a multilingual tool calling to demonstrate how severe observation space gaps lead to operationally invalid actions despite correct semantic intent. Ultimately, this agenda aims to drive a paradigm shift, yielding a unified vocabulary and standardized stress test benchmarks to foster a new generation of highly trustworthy agents for reliable real-world applications.

Xiaoou Liu, Tiejin Chen, Weibo Li +2

Arizona State University Tempe, Arizona, USA

Sim-to-Real Transfer AI Agent Evaluation LLM Agents LLM Agent Reliability

Jun 5, 2026cs.AI

Exploring Agentic Tool-Calling Decisions via Uncertainty-Aligned Reinforcement Learning

Large language model (LLM)-based agents often make suboptimal tool-use decisions, including unsupported tool invocation and hallucinated direct responses, which may accumulate errors throughout multi-step interactions. Existing approaches mainly improve these behaviors through inference-time correction or coarse-grained reward signals based on decision outcomes and structured checklists, leaving the uncertainty characteristics of agent decisions underexplored. We observe that decision-oriented reinforcement learning tends to weaken the uncertainty separation between correct and incorrect actions, resulting in overconfident mistakes and weaker exploration signals. Therefore, we propose TRUST, which incorporates uncertainty quantification into reward design as a repulsive force for maintaining uncertainty separation, and labels lightweight key-turn annotations for unified post-training of multi-turn trajectories. Experimental results across diverse tool-use benchmarks show that TRUST consistently enhances both decision quality and agent performance while maintaining more reliable uncertainty estimates during optimization.

Yijin Zhou, Linqian Zeng, Xiaoya Lu +4

Shanghai Jiao Tong University, China · Shanghai Artificial Intelligence Laboratory, China · Shanghai Innovation Institute, China

Reinforcement Learning LLM Uncertainty Estimation LLM Agent Reliability RL for Tool Use

Jun 4, 2026cs.SE

From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flaws

LLM agents increasingly rely on agent harness: the runtime infrastructure around the base model that defines execution environments, tool interfaces, context, lifecycle orchestration, observability, verification, and governance. Existing self-improving agents and automatic harness evolution methods mainly improve agents through runtime supervision, prompt optimization, workflow search, or harness modification based on final outcomes. However, they often fail to diagnose where the responsible evidence lies in failed trajectories and which harness implementation mechanism causes the unreliable behavior, resulting in broad, indirect, or poorly scoped changes. This paper proposes HarnessFix, a trace-grounded and diagnosis-driven framework for repairing agent harnesses. HarnessFix compiles raw execution traces and harness artifacts into a Harness-aware Trace Intermediate Representation (HTIR), which normalizes fragmented trajectory evidence, captures step-level data-flow and control-flow relations, and aligns runtime steps with the harness artifacts that shape their behavior. It then attributes failures to responsible steps and harness artifacts, and consolidates recurring diagnoses into repair-oriented flaw records. Finally, HarnessFix maps these records to scoped repair operators, generates patches under flaw-specific repair specifications, and accepts them through regression-aware validation. We evaluate HarnessFix on four popular benchmarks, and results show that it improves the performance over the initial harnesses by 6.3% to 18.4%, significantly outperforming human-designed and self-evolution baselines. HarnessFix highlights the value of treating failed trajectories not only as feedback signals, but also as structured evidence for diagnosing and repairing the harness mechanisms behind agent failures.

Mengzhuo Chen, Junjie Wang, Zhe Liu +3

State Key Laboratory of Complex System Modeling and Simulation Technology, Beijing, China · Institute of Software, Chinese Academy of Sciences, Beijing, China · University of Chinese Academy of Sciences, Beijing, China +1

Agent Failure Analysis Self-Improving Agents LLM Agent Harnesses LLM Agent Reliability

Jun 4, 2026cs.AI

ToolChoiceConfusion: Causal Minimal Tool Filtering for Reliable LLM Agents

Large language model agents increasingly rely on external tools, but larger tool menus can reduce reliability and efficiency by increasing wrong-tool calls, premature actions, and token cost. Existing tool-selection methods often optimize semantic relevance, exposing tools whose names or descriptions match the user request. We argue that relevance is insufficient: a tool may be related to the task while still being unnecessary or premature at the current step. We propose Causal Minimal Tool Filtering (CMTF), a training-free method that selects tools by causal sufficiency. CMTF uses lightweight precondition-effect contracts to expose only the minimal next-step tool frontier needed to advance from the current state toward the user goal. Across multi-step tool-use tasks, we compare CMTF with all-tools exposure, keyword retrieval, state-aware filtering, and causal-path ablations, measuring task success, wrong-tool calls, premature actions, tool exposure, and token cost. In the main benchmark with 102 tasks, 100 tools, four LLM backends, and 2448 task-method-model runs, CMTF matches the strongest causal baseline in aggregate success while reducing visible tools from 100 to one per step and reducing token usage by about 90% relative to all-tools exposure.

Rahul Suresh Babu, Laxmipriya Ganesh Iyer

Independent Researcher United States of America

AI Agent Reliability Tool-Use Evaluation Tool-Augmented Language Model Agents LLM Tool Use Tool Selection Tool-Using Language Agents+1

Jun 4, 2026cs.AI

When Tools Fail: Benchmarking Dynamic Replanning and Anomaly Recovery in LLM Agents

Existing benchmarks evaluate Tool-Integrated Reasoning (TIR) in LLMs on idealized ''happy paths'', largely overlooking real-world tool failures. We introduce ToolMaze, a benchmark for dynamic path discovery and error recovery in TIR agents. To separate systematic replanning from blind trial-and-error, ToolMaze adopts a two-dimensional design: DAG-based topological complexity and a

2 \times 2

taxonomy of tool perturbations (explicit/implicit, transient/permanent). Evaluations show that perturbations degrade performance across nearly all models, with the sharpest drops under implicit semantic failures. Driven by systemic over-trust in corrupted outputs, Perturbation Recovery Rate (PRR) plummets by around 37% in these scenarios, while complex topologies trap agents in futile trial-and-error loops. Crucially, agentic fault-tolerance improves with model scale

3.66\times

slower than basic task execution, highlighting dynamic replanning as a distinct bottleneck unaddressed by model scaling or prompting. Data and code are available at https://github.com/Zhudongsheng75/ToolMaze.

Dongsheng Zhu, Xuchen Ma, Yucheng Shen +5

Shanghai AI Laboratory · East China Normal University · Soochow University +2

AI Agent Reliability AI Agent Benchmarks LLM Reliability LLM Tool Use LLM Agent Planning LLM Agent Reliability