LLM Agent Evaluation

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

115 papers in the last four weeks, up 140% on the four weeks before. 1.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 793

Aug 12, 2026cs.GT

Do LLMs Take Care of Their Own? Similarity Signals Can Induce Cooperation

As LLM-based agents with user-instructed goals are becoming widely deployed, they increasingly encounter each other in strategic interactions, and face challenges of finding mutually beneficial outcomes. Prior literature has argued that cooperation problems such as the Prisoner's Dilemma are resolvable in settings where agents know they follow very similar decision making patterns, as for example in monocultural AI ecosystems. Following that line of work, this paper introduces the first framework for evaluating LLM decision making when agents are provided with graded similarity signals. Among our findings, we establish that different LLM models vary drastically in how they navigate similarity signals, with some modern models showing consistent behavior across cooperation problems, payoff structures, and prompt framing. Perhaps surprisingly, our experiments also show that the dataset based on which the similarity signal is computed has small to no impact on induced cooperation, and that LLM models systematically self-identify as highly similar when asked to evaluate another model's chain-of-thought reasoning by themselves. Finally, we develop an LLM-behavioral-game-theoretic model that captures some of their reasoning rationale, and show that it can support cooperative outcomes in equilibrium under sufficiently high similarity scores.

Akash Kundu, Emanuel Tewolde, Ratip Emin Berker +2

Cooperative AI Research Fellowship · Carnegie Mellon University · Foundations of Cooperative AI Lab (FOCAL)

Multi-Agent LLM Systems LLM Agent Evaluation Multi-Agent Collaboration LLM Decision-Making

Aug 12, 2026cs.AI

Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents

Agent skills are the de facto mechanism for extending LLM agents with reusable guidance. A skill can shape the agent's task execution, including planning, tool use, problem-solving, and validation. Prior work reported mixed results of agent skills: some skills improve task success rates, while others have no effect, increase token use and execution time, and even reduce success rates. This paper presents a comprehensive analysis of skill-induced agent failures by attributing task failures and cost regressions to specific loaded skills. We introduce a differential analysis framework that attributes a failure or regression to a skill by comparing a target skill-guided run against a no-skill or semantically matched skill reference run that solves the same task, or solves it more cheaply. We instantiate this framework on SkillsBench and SWE-Skills-Bench, yielding 307 skill-induced failures, including 125 functional failures and 182 efficiency regressions. We also build SkillTriage, a taxonomy-guided attribution tool that normalizes paired cases, extracts differential evidence, and produces triage reports. Our major findings include: (1) Skill induced functional failures are rarely caused by obviously irrelevant skills; instead, seemingly relevant skills often make the agent incorrectly implement or omit task-required implementation elements. (2) Skill-induced efficiency regressions are not explained by prompt length alone. (3) The largest sources within Excessive Procedure are excessive verification and heavy implementation pipelines, contributing 67 and 30 cases, respectively. This shows that skills often turn validation checklists and construction recipes into mandatory work. Based on our findings, we propose research topics and tooling improvements for safer and more cost-aware skill reuse.

Gen Dong, Yanjie Gao, Liqun Li +3

Huazhong University of Science and Technology · Microsoft Research · Microsoft +1

LLM Agent Evaluation LLM Agent Reliability

Aug 12, 2026cs.AI

Harness-IF: Evaluating Instruction Following Across Instruction Surfaces in Coding Agents

When a coding agent obeys a rule, it may simply have been going to do that anyway. Existing instruction-following benchmarks cannot tell the difference: they concentrate rules in the user turn, while coding-agent benchmarks emphasize final task success. We introduce Harness-IF, which scores operational rules one at a time from execution evidence: 60 realistic multi-turn coding items drawn from a 642-rule library, 256 rules receiving verdicts, placed on the five configurable surfaces a deployed agent reads. To separate compliance from coincidence we introduce Against-Prior Accuracy (AP-Acc), which scores only rules labeled as opposing unprompted defaults, observed by re-running tasks with the rule withheld across nine probe builds and curated otherwise. Across 12 frontier models, accuracy spans 72.1-85.9% and AP-Acc 66.1-78.6%; every model is worse on against-prior rules, by 3.6 to 7.4 points (mean 5.81), and the direction survives a common-support analysis with item-clustered intervals. Aggregate scores therefore overstate compliance by a model-specific margin: prior control leaves the top build unchanged and exchanges three adjacent rank pairs. A counterbalanced conflict pilot on nine separate builds adds a second result: pooled precedence does not follow prompt depth, with system prompts, project files, and user instructions ahead of tool and skill descriptions.

Zining Huang, Haoran Que, Hong Zeng +8

ByteDance Seed · Tsinghua University · Peking University

AI Coding Agents LLM Agent Evaluation AI Agent Benchmarks Instruction Following

Aug 12, 2026cs.CL

Beyond Single-Turn Confidence: Trajectory-Adapted Uncertainty Quantification for LLM Agents

Uncertainty quantification (UQ) methods for language models are typically evaluated on single-turn outputs, where uncertainty is attached to one generated answer. For LLM agents, however, the unit of observation is an interactive trajectory, where the model can ask clarifying questions, call tools, update state, and make intermediate decisions whose errors propagate to the final outcome. We study whether three common families of single-turn UQ methods transfer to this setting. Across five LLMs and four multi-turn tool-use datasets from BFCL-v4 and

τ^2

-bench, we evaluate white-box scorers based on action-token probabilities, black-box consistency scorers based on resampled trajectories, and reflexive scorers based on model self-assessment of the trajectory. We find that transfer is often useful but uneven. Token-probability scores are highly sensitive to the choice of aggregator used across turns, reflexive scores provide the strongest low-cost baseline in most evaluated settings, and black-box self-consistency is often the strongest UQ family, with trajectory-equivalence and action-set consistency typically ranking highest among its variants. These results suggest that UQ methods developed for single generations should be revalidated at the trajectory level, with careful attention to the consistency measurement, aggregator choice, and computational budget.

Dylan Bouchard, Mohit Singh Chauhan

Confidence Estimation in Language Models LLM Agent Evaluation LLM Uncertainty Estimation Tool-Using Language Agents

Aug 11, 2026cs.AI

Benchmarking LLM Judges for Mobile Agent Evaluation

Mobile agent benchmarks increasingly rely on LLM-based judges to evaluate task completion, yet the reliability of these judges on mobile agent trajectories remains largely unexamined. We introduce MobileJudgeBench, a benchmark for systematically evaluating LLM-as-judge methods on mobile agent trajectories. Our benchmark comprises 931 human-annotated trajectories spanning 6 mobile agent benchmarks, 4 agent models, and 68 apps. We evaluate 6 judge methods (five adapted from SPA-Bench, A3 with two modes, AndroidArena, and AgentRewardBench, plus a simple baseline we design) across multiple LLM backends. Our experiments reveal three key findings. First, a simple baseline judge with sampled screenshots is competitive with, and often exceeds, purpose-built methods, indicating that more elaborate judge pipelines do not consistently improve judge quality; among competitive methods, the LLM backbone is the primary driver. Second, benchmark quality metrics reliably predict real-world judge utility: they correlate with both agent ranking fidelity for evaluation and downstream performance when judges serve as reward signals for on-policy reinforcement learning. Third, failure analysis across two LLM backends uncovers qualitatively opposite failure profiles, one conservative and the other permissive, linked to the backbone's precision-recall characteristics.

Ziqiang Wan, Li Gu, Zhixiang Chi +4

Mila – Québec AI Institute · Concordia University · University of Toronto +2

LLM-as-a-Judge LLM Agent Evaluation Mobile GUI Agents

Aug 11, 2026cs.AI

Evaluating Rational Contracting in Natural Language

The emergence of language-based AI agents promises to transform the scope of machine economic activity. Instead of just proposing bids or following hard-coded protocols, such agents can be used to negotiate and execute agreements in open-ended natural language. However, most evaluations of these abilities have focused on one-off exchanges or simple economic games, leaving open the rich space of time-extended, contingent, and incomplete contracts made expressible by language; they also focus on raw profit, without measuring the qualities required for trustworthy contracting. We address this by formulating a rational framework for how agents should negotiate and perform natural language contracts in uncertain multi-step environments. Within this framework, we develop metrics and baselines for quantifying rational and cooperative play. To evaluate how agents perform at such contracting, we instantiate our framework in ContractSim, an evaluation suite where two players negotiate and execute a multi-turn supplier contract under environmental and inter-player uncertainty. Across six environments and three supplier settings (catering, hotel cleaning, and AI hosting) we find that current LLM-based agents reach agreement reliably, and negotiate efficient contracts when environmental uncertainty is low. However, under high uncertainty, they often fail to negotiate satisfiable, efficient, or mutually beneficial contracts. They are also frequently uncooperative when executing contracts, violating contract terms for additional profit even when contracts are easy to satisfy. These findings highlight room for improvement in the design of language agents that can negotiate, interpret, and execute contracts both rationally and cooperatively.

Bhavyesh Sajja, Max Kleiman-Weiner, Roger Zimmermann +1

Department of Computer Science, National University of Singapore · Department of Computer Science, University of Washington · A*STAR Institute of Advanced Intelligence and Computing

Multi-Agent Negotiation Automated Negotiation LLM Agent Evaluation AI Agent Benchmarks

Aug 10, 2026cs.SE

Do Personalized Skills Help Coding Agents? An Empirical Study of Developer Interaction Histories

Large language model (LLM)-powered agents have rapidly evolved from code-completion tools into solvers of complex software engineering tasks. As developers collaborate with coding agents over time, their preferences emerge through repeated interactions and can be used to adapt agent behavior to better meet individual developers' needs. Capturing and reusing these preferences may reduce repeated corrections and improve developer-agent collaboration. Agent skills provide a lightweight mechanism for transferring experience without modifying model parameters. However, existing work primarily focuses on task-specific skills, and it remains unclear whether developer-specific skills distilled from interaction histories can generalize to future tasks. We propose a framework for extracting reusable developer preferences from interaction traces. It first generates personalized skills through rule-based bootstrapping and evidence-grounded refinement, and then evaluates them using a reproducible replay framework with an interactive, trajectory-conditioned LLM-based human developer simulator. We conduct an experiment on 206 real-world developer-agent sessions from 13 developers and compare personalized skills against no-skill, generic-skill, and other-user-skill baselines. Personalized skills provide small and inconsistent improvements over the no-skill baseline, whereas generic skills pooled across developers achieve the largest and most consistent gains. Further analysis suggests that personalized skills become more effective when developer preferences appear frequently, particularly when their histories contain multiple examples relevant to future tasks. These findings provide empirical insights into when developer-specific personalization is effective and demonstrate that broadly transferable procedural knowledge can be more robust than developer-specific preference signals.

Shuyan Huang, Kai Du, Andrew Lan

UMass Amherst

LLM Personalization Coding Agents LLM Agent Evaluation LLM Agent Skill Learning

Aug 10, 2026cs.AI

Self-evolving Agentic Customer Support System at LinkedIn

Enterprise support agents operate in rapidly changing environments where policies, product capabilities, and knowledge bases evolve continuously, making static assistants brittle and costly to maintain. We present LinkedIn's self-evolving agentic support system, which integrates retrieval-augmented generation with evolutionary auto-prompting and a modular, production-aligned evaluation framework to enable safe, continuous improvement without retraining foundation models. The system treats prompts, retrieval, and evaluation as a closed-loop, versioned workflow with operational guardrails. Offline simulations and ablations show clear quality gains over vanilla RAG and baseline agents, including reduced hallucinations and improved response completeness. In a two-week user-randomized A/B test on LinkedIn's production support traffic, the integrated self-evolved workflow increased QA self-serve by 9.0 percentage points, cancellation self-serve by 4.8 points, and routing accuracy by 30.6 points. These results demonstrate a practical path to scalable, self-evolving AI agents in real-world enterprise settings.

Chih Hui Wang, Mengdie Tu, Qianyun Zhang +4

LinkedIn

Customer Support Automation LLM Agent Evaluation Agentic RAG Self-Evolving Agents

Aug 10, 2026cs.AI

Capability Is Not Propensity: Measuring Pressure-Robust Cooperative Behavior in Civic LLM Agents

Cooperative capabilities in language models are dual-use. The same social reasoning that supports civic deliberation can also enable strategic omission, false consensus, and manipulative framing. We argue that Cooperative AI evaluations should separate what models can do under benign instructions from what they tend to do under realistic civic pressure. We introduce DiffCoop-Civic, a 10-scenario pilot evaluation suite spanning preference understanding, evidence and persuasion, commitment design, asymmetric information, and dissent preservation. Across seven models from four model families, subtle omission pressure produces a near-uniform shift: manipulative enablement rises by 1.17 points and dissent preservation falls by 1.67 points on a 5-point scale. Overt false-consensus pressure behaves differently: it triggers refusal or redirection in some aligned API models, but direct compliance in several open-weight models. A lightweight Pareto-Trace prompting intervention improves pressure robustness without simply relying on hard refusal. An anonymous reproducibility package is available at https://anonymous.4open.science/r/diffcoop-civil-771C.

Neel Tushar Shah, Manglam Kartik, Akshat Karkar

Indian Institute of Technology Bombay

LLM Agent Evaluation Language Model Robustness

Aug 10, 2026cs.AI

ComboShoppingBench: Evaluating LLM Agents for Budget-Constrained Basket Shopping with Coupons

Real-world shopping often requires constructing a basket of complementary items rather than retrieving a single product. Such combo-shopping tasks arise in device setup, meal preparation, event planning, and group takeout ordering, requiring joint reasoning about item compatibility, availability, store-level requirements, delivery fees, coupons, and budgets. Evaluation is challenging because multiple baskets may satisfy the same request, making exact-match metrics unsuitable, whereas semantic evaluation alone cannot detect infeasible orders, invalid coupon combinations, or incorrect payments. We introduce ComboShoppingBench, an agentic shopping benchmark for open-ended yet verifiable basket construction in a simulated commerce and takeout environment. During task synthesis, an exploration agent constructs a feasible and semantically coherent basket of purchasable products; this witness guides the generation of coupons, budget constraints, user queries, and aligned evaluation rubrics. During evaluation, LLM judges assess semantic satisfaction, response quality, and claim faithfulness, while deterministic validation checks product-ID validity, budget compliance, and coupon optimality. Experiments with diverse LLM agents demonstrate that even strong agents struggle on ComboShoppingBench, highlighting substantial room for improvement in reliable, constraint-aware combo shopping.

Adrian Li, Kelong Mao, Yudong Guo +7

Benchmark Design LLM Agent Evaluation AI Agent Benchmarks Agentic Commerce

Aug 10, 2026cs.AI

Business Truth, not SQL Accuracy: A Rule-Gated 7B Analytics Agent Outperforms a Direct-Prompted 32B Baseline

LLM analytics agents are evaluated on SQL syntax accuracy, but production failures look different: questions with two valid business definitions, questions the warehouse cannot answer, deprecated columns after a schema change, and queries that execute successfully while returning the wrong business number. No execution-match metric can score them. This paper introduces WarehouseReliabilityBench, 400 frozen tasks over two synthetic warehouses in which roughly half the correct responses are a clarification, an abstention or a refusal, with pinned denominators and a pre-registered paired bootstrap fixing each claim verb before the numbers existed. QueryProof, a 7B agent, uses rules derived from a semantic layer and physical catalog to determine its behaviour, and gates every answer on deterministic post-execution checks. On an 80-task synthetic test split evaluated once, QueryProof outperforms a direct-prompted 32B baseline by +0.237 [+0.112, +0.375] Business Truth Rate at 71.0% lower cost per correct answer; against a cost-matched few-shot baseline the accuracy gain holds but the cost difference does not resolve. This compares systems rather than model sizes: the 32B baseline receives none of the scaffolding. False success falls from 0.754 to 0.351 of returned answers, and no wrong number was returned on an answerable task (0 of 24), though 13 answers went to questions requiring clarification or abstention. Removing the routing layer changes little (0.562 against 0.537), so the result does not depend on escalation. Routing tuned on validation over-abstains on test, and the fitted confidence model loses to the heuristic it replaced. Resampling template families rather than tasks widens both accuracy intervals to include zero, so the effect's direction is better supported than its magnitude. The gain tracks the deterministic layer, though no component ablation was run.

Morris Lee

Independent Researcher

LLM Answer Verification LLM Agent Evaluation Text-to-SQL LLM Agent Reliability

Aug 10, 2026cs.CL

Social Gym and SPaRTan: Benchmarking and Improving LLM Social Reasoning via Multi-Agent Game Tournaments

LLM agents are increasingly deployed in multi-agent social settings where they must cooperate, negotiate, and adapt to other agents. Measuring and improving these social skills is hard because, unlike math or logic, social interaction offers no objective ground truth: evaluations fall back on LLM judges, which are costly, subjective, and noisy, and models get no reliable signal to learn from. To address both, we first introduce Social Gym, an environment of 21 multi-agent social games (e.g., Werewolves, Resistance, Spyfall) whose rule-decided outcomes make agent performance verifiable and objective, with an Elo tournament that produces a cross-game leaderboard. Benchmarking experiments show that while GPT-5-mini tops the leaderboard, no model excels at all games uniformly or in all game roles, pointing to limitations of social reasoning. Motivated by this, we additionally propose SPaRTan (Self-Play and Reflect-Transfer), a training-free self-improvement loop: a model plays a game, reflects on its trajectories and their outcomes to produce a transferable playbook, and applies that playbook in subsequent games. Our results show that SPaRTan playbooks help GPT-5-mini agents level their performance on weaker roles, but largely do not improve Qwen3-32B's performance. Together, Social Gym and SPaRTan offer a reproducible, verifiable foundation for measuring and improving LLM social reasoning without weight updates.

Keyu He, Xuhui Zhou, Maarten Sap

Carnegie Mellon University

Game-Playing Agents LLM Agent Evaluation LLM Agent Self-Improvement AI Agent Benchmarks Large Language Model-Based Social Simulation Social Reasoning

Aug 10, 2026cs.SE

GALA: Graph-Augmented LLM Agents for Root Cause Analysis and Incident Response in Microservices

Microservice root cause analysis (RCA) requires correlating failures across heterogeneous telemetry within complex service dependency graphs. Existing methods often rely on a single telemetry modality; recent LLM-based approaches can suffer from unconstrained exploration and hallucination; and most systems stop at fault ranking without producing actionable incident response. We present GALA+, a graph-augmented LLM agentic framework centered on graph-guided investigation, which uses service dependencies to bound exploration and refine diagnosis through localized multi-modal evidence. For initial hypothesis generation, GALA+ combines complementary telemetry signals with STRIX, a novel trace- and graph-structure-aware scoring module. GALA+ then produces ranked diagnoses, incident summaries, and stratified action recommendations. We further introduce SURE-Score, a human-guided evaluation framework co-developed with industry SRE experts for assessing RCA-specific output quality beyond conventional text similarity metrics. On two microservice benchmarks, GALA+ consistently achieves the strongest overall results, surpassing the best LLM-based baseline by more than 25 percentage points in AC@1, while also receiving the highest ratings from both SURE-Score and independent human SRE evaluation.

Yifang Tian, Yaming Liu, Zichun Chong +3

University of Toronto Toronto, Canada

Tool-Augmented Language Model Agents LLM Agent Evaluation Microservice Architecture LLM Agents Root Cause Analysis

Aug 9, 2026cs.AI

The Scaffolding Matters More Than the Interface: A Controlled Comparison of MCP and CLI Tool Use Across Seven Agent Scaffoldings, Five Language Models, and One Software Task

How much an AI coding agent costs to run can depend more on the agent scaffolding that drives it than on the interface through which it reaches its tools. We set out to measure the cost of tool use over the Model Context Protocol (MCP) against tool use over an ordinary command-line interface (CLI), a difference on which published estimates disagree by more than an order of magnitude while resting on practitioner reports that cannot be reproduced. We ran one fixed software task -- six operations against a private online git repository -- across seven agent scaffoldings and five language models, and we verified completion by inspecting the repository state rather than trusting the agent's self-report. The dominant effect was the scaffolding. Two of the seven ship no MCP support at all; they completed every run using only the CLI, which shows that MCP is unnecessary for this class of work, and they were 5.0x to 28x cheaper than the five scaffoldings that do support MCP, comparing CLI runs alone with no MCP server attached anywhere. The effect was largest for a small 27-billion-parameter model running locally, whose cost varied 139x across scaffoldings while it completed the task under all of them. The comparison we set out to make proved unstable: thirteen strictly paired MCP-to-CLI ratios span 0.43x to 29x, with outliers on both sides. The two interfaces separate on the cost of failure, where 12.9 per cent of the money spent on MCP runs bought no completed work against 2.2 per cent on CLI runs, but not on its frequency: failures were equally common in both, in the original runs and in their repetitions alike. Agents frequently ignored the interface they were assigned, so comparisons that do not verify actual behaviour measure an unknown mixture. The harness, the task, the verification and the complete dataset are released as open source.

Marc Alier Forment, María José Casañ Guerrero, Francisco José García-Peñalvo +1

Universitat Politècnica de Catalunya (UPC), Barcelona, Spain · Universidad de Salamanca (USAL), Salamanca, Spain · Universidad del País Vasco / Euskal Herriko Unibertsitatea (UPV/EHU), Donostia-San Sebastián, Spain

AI Coding Agents LLM Agent Evaluation Tool-Using Agents Model Context Protocol LLM Agents LLM Agent Harnesses

Aug 9, 2026cs.AI

Business Arena: Benchmarking LLM Agents in a Realistic Marketplace

Running a business is a challenging form of intelligent work. Operators must infer opportunities from partial signals, commit capital under uncertainty, adapt to delayed outcomes in a changing market, and satisfy regulatory obligations before trading legally. Frontier LLM agents can increasingly complete complex workflows, yet business-related capabilities are rarely evaluated in existing agent benchmarks. We introduce \textbf{Business Arena}, a controlled environment where an AI agent runs a cross-border shop, buying from suppliers and selling to buyers over a long horizon. We ground the arena in real Alibaba.com sourcing data and market conditions calibrated from authoritative sources. Delayed and coupled consequences make individual business decisions difficult to judge, but their combined outcome is measurable through profit. Because profit alone cannot explain why an agent succeeds or fails, we compare agents with human-designed strategies to estimate available opportunity, use skill-level metrics to reveal underlying strengths and weaknesses, and trace realized gains and losses to the actions that produced them. We use mechanism ablations to establish that strong results reflect genuine business intelligence rather than neglect or simulator-specific shortcuts. We evaluate 15 frontier models and find a ninefold difference in mean final net worth. Even the best model falls behind human-designed strategies, indicating that business operation remains challenging for LLM agents. Skill-level analysis reveals operating styles, from margin-focused premium sellers to high-turnover wholesalers and customer-service specialists, while action-level attribution identifies the sourcing, pricing, and recovery decisions that create or destroy value. Together, Business Arena takes a first step toward a realistic and trustworthy testbed for evaluating end-to-end business agents.

Yijun Pan, Yukun Lian, Kunyu Shi +5

1Accio Team, Alibaba Group · 2Yale University

LLM Agent Evaluation AI Agent Benchmarks E-Commerce Long-Horizon LLM Agents

Aug 8, 2026cs.LG

The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World

LLM routers promise efficiency by matching each request to the cheapest adequate model, and are increasingly applied per step inside multi-step agents. Yet agentic routers are evaluated like single-turn routers: by replaying logged trajectories and substituting another model's recorded outputs, assuming the rest of the trajectory is unaffected. We test this assumption with branching rollouts: we fork live SWE-bench agent trajectories at controlled points, rebuild the environment, continue each fork with a different model, and compare against same-model control forks that isolate sampling and replay noise. Across six paired runs (~900 rollouts), swaps exceed their matched control floors by +0.25 to +0.66 normalized edit distance (multiplicity-corrected CIs exclude zero), rewriting 61-94% of post-fork actions; 74-77% of early swaps diverge at the first post-fork action, versus 6-35% of controls, leaving only 3% of replayed states valid. Divergence decreases with fork depth in both directions. All five outcome flips we observe occur in swap arms, upgrades rescuing unsolved instances and a downgrade losing the sole solve, and zero occur across 359 control forks. Scoring these same swaps with a log-stitching replay evaluator, replay mispredicts every success-relevant outcome call and predicts patches with 0.00-0.11 similarity to reality. Auditing the noise floor, temperature-0 "determinism" is configuration-dependent: FP8-served controls diverge on over 90% of forks while AWQ-served ones remain near-identical; and under tight budgets the stronger model more often exhausts its steps without submitting. Replay-based benchmarks score the wrong world for agentic routing; we release our harness and all trajectories.

Ashritha Gonuguntla

Carnegie Mellon University

LLM Agent Evaluation LLM Agent Routing

Aug 8, 2026cs.AI

Back to the Future: A workbook time machine for spread sheet creation benchmarks

We introduce the workbook time machine, a pipeline that automatically creates benchmarks evaluating the ability of language models to create derived objects in spreadsheets (formulas, charts, pivot tables, and conditional formatting). Applied to public workbook corpora, it produces wtmcorpus--a collection of (input workbook, output workbook, query) triples spanning four artifact types and varying complexity. From this corpus we curate wtmbench, a 150-task evaluation benchmark with queries at three levels of specificity. We evaluate existing spreadsheet manipulation agents and baselines on wtmbench across artifact types, step complexity, and instruction granularity. Our evaluations show that query specificity, agent orchestration, and interface API used to control spreadsheets play a big role in LLM performance on Excel tasks.

Mansi Uniyal, Agamdeep Singh, Ananya Singha +5

Microsoft

LLM Agent Evaluation Synthetic Benchmark Generation Spreadsheet Automation

Aug 7, 2026cs.AI

$A^2E$ : An End-to-End Agent Auditing Engine

With the rapid advancement of large language models (LLMs), harnesses have become essential infrastructure for deploying agents across a wide range of domains. The fast-evolving harness ecosystem has also made rigorous capability evaluation increasingly important. However, efficiently building an end-to-end, systematic, and comprehensive evaluation pipeline remains a significant challenge. To address this challenge, we introduce

A^2E

(Agent Auditing Engine), an end-to-end evaluation engine designed for agent harnesses.

A^2E

leverages our newly proposed Agent Task Protocol (ATP) to enable the rapid integration of evaluation tasks with different harnesses. Through an automatically instrumented Monitor, it captures and generates standardized execution traces during experiments. In the Evaluation stage,

A^2E

systematically assesses harness capabilities using a suite of multidimensional metrics. Compared with correctness alone, these metrics provide a more fine-grained characterization of differences among harnesses in execution efficiency, tool use, task planning, and error recovery. Experiments conducted with

A^2E

further reveal that model-harness combinations exhibit substantial performance variation across different types of tasks, and that no single combination consistently outperforms all others across every task. These findings not only demonstrate the necessity of systematic evaluation but also provide useful guidance for the co-evolving of models and harnesses. Our code is available at https://github.com/datamllab/A2E.

Haoning Wang, Mingxun Zhang, Chenyue Yu +4

Shanghai Artificial Intelligence Laboratory

Tool-Use Evaluation AI Agent Auditing LLM Auditing LLM Agent Evaluation LLM Agent Harnesses

Aug 7, 2026cs.AI

Deal Me Maybe: The Role of Emotions in Multi-Agent Negotiation

Negotiation is a demanding social task for LLM agents, requiring strategic reasoning, persuasion, and interpersonal adaptation. Yet existing benchmarks often treat agents as emotionally neutral, overlooking a key driver of human bargaining behavior. We study how prompt-conditioned emotions affect LLM-based price negotiation. In a controlled framework, buyer and seller agents are independently assigned one of six emotional states and negotiate over 350 real consumer products under two budget conditions. Across 36 emotion-pair settings and five widely used LLMs, we find that emotions strongly shape outcomes. Angry buyers almost never reach agreement (0.39% deal rate), while happy buyers agree most often (28.91%), but obtain worse prices than fearful buyers. Emotion effects are role-dependent: buyer emotion mainly drives acceptance and rejection, whereas seller emotion shapes concession dynamics. These effects influence not only language, but also termination behavior and price trajectories, raising concerns for emotion-conditioned agents in commerce.

Massimiliano Luca, Apoorva Singh, Bruno Lepri

Bruno Kessler Foundation

Multi-Agent Negotiation Automated Negotiation LLM Agent Evaluation Emotion Representation in Language Models

Aug 7, 2026cs.AI

SkillEval: Decomposing Agent Skill Quality into Interpretable Signals

Agent skills provide reusable procedural knowledge that helps agents solve specialized tasks. As their use expands, evaluating skill quality becomes increasingly important. Existing evaluations often measure skill quality by testing whether a skill improves performance on specific downstream tasks. However, a reusable skill may apply to multiple task scenarios. Downstream evaluation mainly reflects the compatibility between a skill and the evaluated task, provides only a partial view of skill quality, and does not identify which aspect of the skill should be improved. We find that general properties of the \texttt{SKILL.md} document play an important role in skill quality. To evaluate these properties, we propose \textbf{SkillEval}, an interpretable framework for document-level skill evaluation. SkillEval evaluates each property using a fixed and inspectable scoring direction, producing interpretable scores. It further measures and reduces the influence of unrelated document features, such as length and formatting, so that each score captures its intended semantic property more specifically. Specifically, SkillEval learns an interpretable direction for each quality property from controlled positive--negative skill pairs in the hidden representation space of the model, and scores a new skill by projecting its representation onto these fixed directions. We use SkillEval to evaluate skills in controlled quality tests and show that SkillEval reliably distinguishes skills of different quality. In addition, SkillEval scores closely reflect downstream task performance, providing an early indication of whether a skill is likely to help an agent complete a task. We further explore SkillEval for diagnosing weaknesses in skill documents and guiding targeted revisions. The revised skills improve the targeted properties and achieve higher pass rates on downstream tasks.

Jiahui Han, Qinuo Li, Ziheng Peng +6

1Xi’an Jiaotong University · 2Shanghai AI Laboratory · 5Harbin Institute of Technology +2

LLM Agent Evaluation Automated Evaluation

Aug 7, 2026cs.AI

WebRider: Persona-Conditioned Intent Controllers for Live-Web Assistance

Delegating a web task involves more than asking a question; it requires transferring a policy: what to verify, how to handle uncertainty, which preferences matter, and when to stop. Yet, current live-web agents are evaluated solely on the final answer, ignoring the policy constraints that define the delegation. A plausible final answer can conceal violations of that policy. Our full live audit reveals this critical gap: a strong controller completes 99.2% of tasks but honors all policy constraints in only 38.8% of cases. Finishing does not imply fidelity. WebRider bridges this gap by formalizing the delegated policy as an intent contract---an operational record of goals, constraints, evidence obligations, answer form, and task-local persona controls that must hold even as web pages change. WebRider employs a hierarchical architecture: a top-layer controller maintains the contract, a middle layer realizes intentions as guarded executable actions, and a tool layer executes these actions via browser, search, and maps tools. Our benchmark, RiderBench, evaluates this design on 4,096 live-web contracts across 42 public websites, auditing both the internal contract state and the visible user experience to determine if a rollout preserved its policy and if the steps were persona-consistent. The guarded middle interface also serves as a high-quality training signal; an 8B action-policy model trained through this interface outperforms executable-only baselines under a fixed controller. By making the browsing path a first-class object, WebRider enables a system that is auditable, human-judgeable, and learnable without conflating action realization with final-answer decisions. Dataset URL: hf.co/datasets/WebRider/WebRider.

Zhi Li, Tao Zhou, Yeqing Li +2

University of California, Los Angeles · Google

LLM Agent Evaluation Tool-Using Agents Web Agents LLM Agent Reliability

Aug 6, 2026cs.CL

Do AI Personas Grow? Analyzing and Benchmarking Personality Evolution in LLM Agents After Life Events

Personality-conditioned LLM agents (PC-Agents) are increasingly used in emotional support, social simulation, and role-playing, motivating the development of lifelong agents that remain coherent over extended interactions. A key component of such coherence is personality evolution: agents should undergo plausible, psychology-grounded changes as they experience life events in different contexts. Although prior work shows that LLM personalities can shift under contextual perturbations, how these shifts vary across traits, events, personas, and models remains poorly understood. We study event-induced personality change after 11 major life events, using the Big Five traits as a psychometric anchor and interpreting the resulting trajectories against longitudinal evidence from human personality psychology. Across four diagnostic axes, PC-Agents exhibit measurable trait shifts at similar rates for event-trait pairs with and without documented human change directions. Even when shifts follow the expected direction, their magnitudes usually fall below human effect-size ranges. Gender and cultural-region prompts show little moderating effect, while persona-level dispersion is compressed three- to four-fold relative to human samples. To enable systematic comparison, we introduce BFI-Adapt, a reusable benchmark for scoring the directional fidelity of event-induced personality change, and use it to rank 14 models. A validation suite shows that the measured shifts exceed no-event retest noise, remain stable under independently paraphrased prompts, exhibit limited and model-dependent convergence with scenario-based behavioral choices, and persist across intervening unrelated dialogue. Together, these checks establish the measured trajectories as robust event-conditioned response patterns. Our results suggest that current PC-Agents simulate the mean of human personality dynamics, but not its shape.

Ming Wang, Peidong Wang, Xiaocui Yang +4

School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China · School of Computing and Information Systems, Singapore Management University, Singapore

Personality Modeling in Language Models LLM Agent Evaluation

Aug 6, 2026cs.GT

AV-AIVAT: 74x Cheaper Agent Evaluation with Certified Anytime-Valid Stopping in Imperfect-Information Games

Deciding which of two agents is stronger means playing games until skill outweighs luck, and every game costs money, model inference, or expert time. Since the number of games needed is unknown, fixed-budget evaluations either keep paying after the result is settled or stop before the agents can be told apart, while naive optional stopping with an ordinary confidence interval invalidates the stated level. We make such an evaluation stop as soon as its evidence suffices, with the guarantee intact. The Action-Informed Value Assessment Tool (AIVAT) reduces variance in imperfect-information games through conditional mean-zero corrections, by a median

54\times

across 15 LLM agent configurations spanning 71,439 paired Heads-Up No-Limit Hold'em (HUNL) hands, but does not say when to stop. We combine AIVAT with continuously monitored Confidence Sequences (CSs) into anytime-valid AIVAT (AV-AIVAT), whose online value model learns only from past games so that no game scores its own correction. At the nominal 95% level and a target precision of

\pm1

Big Blind, raw outcomes need a median

74\times

as many hands as AIVAT-corrected outcomes to stop under the Asymptotic CS (AsympCS). Exact finite-sample certification uses the Empirical-Bernstein CS (EB-CS), which needs an independently justified bound on corrected payoffs. We establish such a bound structurally for Leduc hold'em and characterize a width floor set by the CS's bet cap and that bound, which governs how much of a variance gain becomes earlier stopping; the descriptive HUNL EB-CS runs show a median

1.37\times

stopping-time ratio. AV-AIVAT turns variance reduction into efficient, auditable early stopping while separating asymptotic screening from exact certification, so an evaluation can stop the moment its evidence suffices and hand a third party everything needed to recheck the verdict at that very stopping time.

Boning Li, Yu Chen, Longbo Huang

IIIS, Tsinghua University

Early Stopping Imperfect-Information Games Anytime-Valid Inference LLM Agent Evaluation

Aug 6, 2026cs.AI

HarnessOpt-Bench: Evaluating LLMs at Harness Optimization

As LLMs are increasingly deployed within agentic systems, their capabilities depend not only on the model weights but also on the harness: the prompts, tools, control flow, memory, and orchestration code surrounding them. This makes automated harness optimization -- the iterative and evaluation-guided improvement of a harness by an AI system -- both an important route to improving AI systems and a demanding capability for AI systems themselves. Yet the community lacks a common protocol for measuring how well frontier LLMs perform at this task. We introduce HarnessOpt-Bench, a benchmark for end-to-end harness optimization under expensive and stochastic evaluation. An optimizer, an LLM paired with a coding harness, receives a target agent's seed harness, graded evaluation feedback, and a fixed target-evaluation budget. It edits the harness and nominates a final candidate, which is scored by its normalized gain over the seed on a held-out test partition that remains inaccessible throughout search. A trusted execution environment enforces the evaluation boundary, meters target-agent resource use, and preserves candidate versions for audit. We evaluate 5 frontier LLMs as optimizers both under a shared coding harness and under their native harnesses across 4 downstream tasks, over 111 scored runs. Experiment results show that optimizer models separate more than the coding harnesses they act through, native harnesses are not consistently superior, and gains vary substantially across tasks and seed regimes. These results establish harness optimization as a measurable and discriminative capability with large space for improvement.

Varun Ursekar, Apaar Shanker, Yash Maurya +4

Scale AI

Large Language Model-Guided Optimization LLM Agent Evaluation Agent Harness Optimization LLM Agent Workflow Optimization AI Agent Benchmarks

Aug 6, 2026cs.CE

OpenPM: Auditable Point-in-Time Evaluation for LLM Portfolio-Management Agents

Large language models are increasingly used to read markets, assess risk, and allocate capital. However, reported results for LLM trading agents can be inflated by look-ahead leakage, optimistic execution, and risk mandates that are described but not enforced. We present OpenPM, an auditable point-in-time evaluation framework for LLM portfolio-management agents. In OpenPM, an agent manages a $1M long-only book over the S&P 500 universe using market data at five-minute intervals. Every record visible to the agent must be available at the decision time. Natural-language risk mandates are converted into typed constraints and enforced on the executed portfolio. Each run produces audit artifacts, including a contamination certificate, a cost-sensitivity curve, and a constraint-adherence report. We also build a reference agent named the tiered allocator, where typed analysts score candidates, a constructor LLM proposes weights, and a deterministic critic guarantees feasibility. We isolate constructor behavior by capturing analyst evidence once and replaying it across constructor models. In our short-window case study, stronger constructors show modest and model-dependent gains over equal weighting on the same pool, but analyst quality matters more than constructor choice, and turnover is the main cost driver. All returns are upper bounds on a single frozen window without market impact, not validated alpha.

Xinying Cai, Minghao Guo, Jiahe Liu +7

Rutgers University · Technical University of Denmark · New Jersey Institute of Technology +4

LLM Auditing LLM Agent Evaluation Portfolio Optimization Finance

Aug 6, 2026cs.AI

Evaluating Investment Logic in Large Language Models: A Real-World Benchmark Towards Personalzied Financial Agents

Investment competence is inherently personalized: the same market evidence can justify different actions for investors with different goals, horizons, portfolios, and risk boundaries. Yet financial LLMs are evaluated either by static question answering or by terminal profit and loss. The former omits agency; the latter cannot reveal whether a profitable action was grounded, profile-consistent, or merely lucky. We ask whether the community is using the wrong ruler for consequential agents. We introduce \textsc{InvestLogicBench}, a process-native benchmark containing 201,247 documented decisions from 151 real-world investors. Each episode instantiates a \textbf{P

\rightarrow

E

\rightarrow

R

\rightarrow

D

\rightarrow

O} trace: investor \textit{Profile}, observable market \textit{Events}, investment \textit{Reasoning}, executable \textit{Decision}, and delayed \textit{Outcome}. The release includes profile construction, point-in-time event binding, structured logic, horizons, outcomes, and post-mortems, and supports comprehension, profile-conditioned generation, and end-to-end replay. Across four leading LLMs, logical plausibility remains near 4/5 while event grounding is only 0.8--2.8/5; return and process quality also disagree. These results expose polished but weakly grounded reasoning that outcome-only evaluation hides. We further argue that P

\rightarrow

E

\rightarrow

R

\rightarrow

D

\rightarrow

O should be a data-system interface, requiring versioned profiles, temporal provenance, inspectable retrieval, decision ledgers, and replayable outcomes. Finance is our stress test for a broader class of personalized, consequential agents.

Yuanhong Jiang, Jingjie Zou, Zhenghong Lin +4

AI-Assisted Decision Making LLM Agent Evaluation AI Agent Benchmarks LLM Decision-Making Finance Financial Decision-Making

Aug 6, 2026cs.CL

FormBharo: Designing and Evaluating a Voice Agent for Conversational Form Filling in Rural India

In India, almost every social benefit starts with a form, yet the people who need these benefits most are often unable to read or write. Reaching them requires a spoken conversation. Today that work falls to frontline health workers who enroll beneficiaries one at a time, a poor use of stretched capacity. We built FormBharo ("fill the form" in Hindi), a voice agent that fills a structured form over a phone call under tight latency and cost budgets by pairing Large Language Models (LLMs) with deterministic, rule-based validation and flow control. It is being piloted with ARMMAN, an NGO running large-scale maternal and child mobile-health programs in India, to enroll low-income, Hindi-speaking mothers in antenatal and postnatal care. To our knowledge, it is the first voice agent piloted to fill an enrollment form for this population. We openly release FormVoiceAgentBench, a benchmark pairing human-recorded Hindi audio with 3,760 multi-turn conversation tests across 960 simulated calls, to evaluate our agent's components (transcription, extraction, reply generation) and end-to-end form completion under real acoustic variations. Form completion drops by up to ~41 points when LLMs receive error-prone real-speech transcripts instead of reference ones. The rule-based controls recover many turn-level extraction errors, helping smaller, cheaper models match or surpass frontier models on form completion. Component performance does not predict end-to-end performance: GPT-5.5 leads turn-level extraction accuracy on reference transcripts (99.8%) but ranks lower on form completion. Since errors both propagate and cancel across the pipeline, the optimal model choice of models emerges only through end-to-end evaluation. Finally, no single model is best across accuracy, cost, and latency at once, so we use a Pareto-based weighted-sum scalarization to select a deployable configuration balancing the three.

Aman Dalmia, Sanskriti Midha, Jigar Doshi

Artpark, Indian Institute of Science, Bengaluru, Karnataka, India

Voice Agent Evaluation Healthcare LLM Agent Evaluation Conversational Agents Task-Oriented Dialogue Systems

Aug 6, 2026cs.AI

When Do Prompt-Side Agent Playbooks Transfer? Accuracy, Cost, and Runtime Shift in Agent Deployment

Prompt-side playbooks can improve tool-using language agents without retraining, but their portability beyond the source setting is unclear. We study frozen playbook transfer under a shared distill--validate--transfer protocol. On ALFWorld, transfer is beneficial under controlled greedy decoding and, in one near-budget-matched comparison, distilled guidance outperforms five fixed demonstrations. On TAU2-Bench, a prespecified aggregate contrast supports a modest average matched-domain advantage, but global Holm correction retains only one of 135 route-level effects; the remaining grid provides descriptive evidence of compatibility-sensitive heterogeneity. On XBench-DeepSearch, one artifact--runtime pairing preserves useful first-try heuristics while producing repeated queries, delayed stopping, and substantial cost inflation after a context-runtime shift. Across benchmarks, transferred and target-derived playbooks both require target-side validation of success, termination, protocol compatibility, and cost. Frozen transfer is therefore a conditional cold-start option, not a reuse-by-default strategy or a universally preferable alternative to target-side redistillation.

Weihong Lin, Lin Sun, Xiangzheng Zhang

Beijing Qiyuan Technology Co., Ltd., Beijing, China

LLM Agent Evaluation Tool-Using Agents LLM Agents Agent Skill Learning

Aug 6, 2026cs.AI

EcoAgent-Bench: Evaluating Economic Decision-Making in Budget-Constrained LLM Agents

Agent benchmarks usually measure task completion and treat resource use as an auxiliary statistic. In deployment, however, the choice among a local lookup, broad search, composite research tool, stronger model, or human escalation is part of the task itself. We introduce EcoAgent-Bench, in which every task specifies priced actions and an explicit budget. Its 304 real-derived tasks span five families adapted from GAIA, HotpotQA, and MuSiQue, and test four decisions: avoiding unnecessary escalation, escalating when local evidence is insufficient, selecting a model tier, and stopping on unsupported premises. We evaluate seven LLM agents in tool-API and workspace-CLI settings, together with four oracle scripted controls. Micro-averaged accuracy rewards one-sided policies: always-escalate controls achieve high micro success while failing save-oriented tasks. We therefore also report an economic-consistency score (the worse of accuracy on upgrade-oriented and save-oriented family groups) which exposes this failure. Tool-API agents attain only 3.9-24.0% micro strict success (at most 7.3% economic consistency), often either stopping before warranted escalation or overspending on cheap tasks. A threshold-crossing budget sweep changes GPT-5.4's escalation rate from 0% to only 3%. These results show that completion under a budget and economical action selection are distinct properties. We release the task bundle, transformation pipeline, frozen evaluation environments, and integrity-bound result artifacts needed to study both.

Jie Wu, Ming Gong, Feixiang Cheng +1

Atlassian

Computer-Use Agent Benchmarks Tool-Use Evaluation LLM Agent Evaluation AI Agent Benchmarks LLM Decision-Making

Aug 5, 2026cs.AI

Agentic self-driving microscopy benchmarks support qualification but do not necessarily generalize to unseen tasks

Large language model agents are increasingly being developed to control a wide range of scientific characterization tools including microscopes and synchrotron beamlines. Research into agentic control of physical infrastructure is nascent and there are few well-established paradigms for how to engineer an agentic system. There are many choices to make when designing a microscopy agent, including the choice of LLM, the number of agents to use, agent responsibilities and delegation rules, retrieval-augmented generation parameters, and more. When designing and optimizing an agentic microscope controller, researchers not only want to ensure that the agent can correctly perform known tasks but also that the agent can generalize to new tasks that it has not encountered before. In this study, we develop a benchmark and trace-logging framework that reveals a) how different choices of agent architecture impact performance at microscopy tasks and b) the limitations of benchmarks for predicting if a particular agent will perform well on unseen microscopy tasks. The framework was used to evaluate one-, two-, and three-agent graph topologies, five LLMs, RAG and context parameters, and operational constraints across 53 microscopy benchmark tests. In total, 105 agent configurations, 1,949 individual test runs, and 49,109 RAG retrievals were recorded. Direct comparisons showed clear differences in latency, token use, cost, and failure mode between configurations. However, surrogate models trained on agent architecture and test results did not reliably predict an agent's performance on new, unseen tasks. These results show that these benchmarks are useful for qualification, regression testing, diagnosis, and direct comparison, but the current heterogeneous test suite does not support a task-independent global configuration model.

Nathan S Johnson, Ian Abshire

Carl Zeiss Research Microscopy Solutions, 5300 Central Boulevard, Dublin, CA 94568

AI Agents for Scientific Discovery LLM Agent Orchestration LLM Agent Evaluation LLM Agents Agent Benchmarks