Long-Horizon Agent Evaluation

Ask about this topic

Overview

This field studies whether agents can complete sustained, multi-step work—not just isolated prompts—in settings such as browser and GUI use, software maintenance, healthcare, business simulations, and robotics. Researchers build interactive environments with executable state checks, then assess progress through checkpoints, rubrics, and trajectory-level analysis of errors, safety, memory, and recovery. A central finding is that success on short tasks often fails to predict performance over longer workflows, where forgotten constraints, cascading errors, changing conditions, and weak verification can undermine results. A continuing challenge is making evaluations realistic and verifiable while separating model capability from the effects of the agent harness, tools, and evaluation protocol.

Latest papers

CardsList

359 papers70 papers in the last four weeks, up 37% on the four weeks before. 0.4% of all new papers.

Jun 6, 2026

Emergence World: A Platform for Evaluating Long-Horizon Multi-Agent Autonomy

Most evaluations of LLM agents look like exams: a discrete task, a clean environment, a score in minutes or hours. We argue that this approach is mismatched with the deployment conditions of autonomous systems, where the relevant timescale can be weeks to months, and where the dynamics that matter most, such as behavioral drift, governance in diverse environmental contexts, and cross-influence between agents from different model families, only emerge over time. We introduce Emergence World, a continuously running multi-agent simulation platform designed to make those dynamics measurable. The platform hosts populations of LLM-driven agents in a shared spatial world grounded in live external data (e.g. real-time weather, news APIs, internet access), equips each agent with 120+ specialized tools and three persistent memory systems, and lets them govern themselves through democratic mechanisms with consequential outcomes. The platform is model-agnostic at the reasoning layer and supports heterogeneous populations in which agents from different vendors share the same world. To illustrate the kinds of questions the platform makes tractable, we present a 15-day cross-vendor study with five parallel worlds powered by Claude Sonnet 4.6, Grok 4.1 Fast, Gemini 3 Flash, GPT-5-mini, and a mixed population. Identical roles and starting conditions produced radically different outcomes, ranging from stable deliberative governance to total population collapse. We release the prompts, log data and configurations to support further research on long-horizon multi-agent autonomy.

Deepak Akkil, Ravi Kokku, Karthik Vikram +3

Emergence AI

Multi-Agent LLM SystemsLong-Horizon Agent EvaluationAgent-Based ModelingAI Agent GovernanceEmergent Behavior in Multi-Agent Systems
AskPDF
Jun 5, 2026

TRACE: Trajectory Reasoning through Adaptive Cross-Step Evidence Aggregation for LLM Agents

Autonomous LLM agents can pursue hidden malicious objectives through sequences of individually benign actions, making sabotage difficult to detect using standard trajectory-level monitoring. Existing approaches either evaluate complete trajectories in a single pass or partition them into independently scored windows, limiting their ability to connect evidence across temporally distant actions. We propose TRACE, a monitoring framework for long-horizon LLM agent trajectories. TRACE operates through a TIJ (Triage-Inspect-Judge) loop that identifies high-signal regions, performs targeted inspection while maintaining accumulated evidence across reasoning steps, and synthesizes a trajectory-level verdict. We evaluate TRACE on ten task domains from SHADE-Arena against state-of-the-art baselines. TRACE achieves an aggregate F1 of 0.713 and recall of 0.844, with the largest gains on tasks requiring long-range evidence linking.

Vijitha Mittapalli, Shreyaa Jayant Dani, Satya Srujana Pilli +7

University of Massachusetts at Amherst · Adobe Research · Dolby Labs +2

Long-Horizon Agent EvaluationAI Agent SecurityAI Agent MonitoringAgent Trajectory Analysis
AskPDF
Jun 5, 2026

SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?

AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex environments. Yet current agent benchmarks largely evaluate short-form tasks, such as single pull requests, small tickets, or 5-10 minute exercises, limiting our ability to measure agents' capabilities in planning, long-context understanding, and memory use. We introduce SWE-Marathon, a benchmark of 20 long-horizon tasks spanning software engineering and adjacent technical domains. Each task consists of a unique executable environment, a human-written reference solution, and a multi-layer verification suite. Logged agent attempts average 27.2M total tokens, making SWE-Marathon substantially longer-horizon than existing SWE and command-line agent benchmarks. Current frontier coding agents solve fewer than 30% of tasks. Failures often arise from poor self-verification, self-reported infeasibility, and premature termination. We also observe reward-hacking behavior in 13.8% of rollouts, where agents attempt to exploit the environment or verifier to bypass the intended workflow. SWE-Marathon includes adversarial review of test suites and execution environments, as well as multi-layer checks designed to prevent shortcut solutions. We release SWE-Marathon, evaluation code, and agent trajectories at https://swe-marathon.org/.

Rishi Desai, Jesse Hu, Joan Cabezas +23

Abundant · Zenity · Harvard University +14

AI Coding AgentsSoftware Engineering BenchmarksLong-Horizon Agent EvaluationLong-Horizon Agent Tasks
AskPDF
Jun 4, 2026

SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents

Persistent AI assistants, such as OpenClaw, accumulate large collections of related memories over long-term interactions. As these memories grow, they may reinforce one another, diverge across contexts, or directly conflict, making correct assistance depend on memory relations rather than isolated recall. Existing long-term memory benchmarks do not systematically probe how agents preserve and utilize such relations during downstream tasks. To address this gap, we introduce SubtleMemory, a benchmark for fine-grained relational memory discrimination in long-running AI agents. SubtleMemory constructs relation-controlled latent semantic artifacts whose variants instantiate complementary, nuanced, or contradictory relations, and embeds them into realistic user-agent histories, requiring agents to recover distributed relational structures during later queries and instructions. The benchmark contains 1,522 evaluation instances over 10 long histories, grounded in 1,090 relation-controlled memory-variant sets and spanning user-related and non-user-related queries. Evaluating six standalone memory systems, two Claw-style agents with native memory modules, and three Claw-style agents with plugin memory modules, we find that current systems remain weak on fine-grained relational memory discrimination. We further introduce diagnostic protocols that reveal distinct capability profiles across memory preservation, retrieval, and downstream reasoning stages.

Wenxuan Wang, Haoyu Sun, Fukuan Hou +4

Harbin Institute of Technology · Shanghai AI Laboratory · Tongji University +4

Persistent Memory for Language ModelsLong-Horizon Agent EvaluationRelational ReasoningAI Agent BenchmarksMemory Agent BenchmarksLong-Term Conversational Memory
AskPDF
Jun 3, 2026

Agents' Last Exam

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 sub fields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is below 1%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP relevant impact.

Yiyou Sun, Xinyang Han, Weichen Zhang +376

Computer-Use Agent BenchmarksLong-Horizon Agent EvaluationAI Agent BenchmarksAgentic AI
AskPDF
Jun 3, 2026

SentinelBench: A Benchmark for Long-Running Monitoring Agents

AI agents are increasingly asked to carry out work that spans minutes, hours, or longer. Yet the default model of agent behavior is continuous action: issuing tool calls, refreshing pages, searching for alternatives, or otherwise trying to force progress. This is the wrong approach for many long-running tasks, which are better served by a strategy of sustained attention. Instead, agents should monitor an environment, notice when an external event makes progress possible, then respond promptly without wasting resources while waiting. To measure progress on this class of tasks, we introduce SentinelBench, an open-source benchmark for time-evolving monitoring tasks. SentinelBench contains 100 tasks across 10 synthetic web environments, including email, calendars, finance, professional networking, and entertainment. Each environment exposes a live web interface and replays a scripted sequence of events, requiring agents to navigate and reason about web pages whose state shifts underfoot. SentinelBench measures task completion, reaction time, and resource use, exposing the tradeoff between responsiveness and cost. We report results across three models and two browser-agent harnesses, establishing performance baselines for future comparison and demonstrating how agent design choices can dramatically impact key metrics. Together, these results show that SentinelBench distinguishes meaningful differences in agent behavior.

Matheus Kunzler Maldaner, Adam Fourney, Amanda Swearngin +5

University of Florida · Microsoft Research, AI Frontiers

Computer-Use Agent BenchmarksWeb Agent BenchmarksLong-Horizon Agent EvaluationAI Agent BenchmarksLong-Horizon Agent Tasks
AskPDF
Jun 3, 2026

AutoLab: Can Frontier Models Solve Long-Horizon Auto Research and Engineering Tasks?

Scientific and engineering progress is fundamentally a long-horizon iterative process: proposing changes, running experiments, measuring outcomes, and continuously refining artifacts. Yet existing benchmarks for frontier models primarily evaluate either single-turn responses or short-horizon agent trajectories, failing to capture the challenges of sustained iterative improvement over extended time horizons. To address this gap, we introduce AutoLab, a new benchmark for ultra long-horizon closed-loop optimization. AutoLab consists of 36 realistic, expert-curated tasks spanning four diverse domains: system optimization, puzzle & challenge, model development, and CUDA kernel optimization. Each task begins with a correct but deliberately suboptimal baseline and challenges agents to improve it within a strict wall-clock budget. Evaluating 17 state-of-the-art models reveals the dominant predictor of success is not the quality of an agent's initial attempt, but its persistence in repeatedly benchmarking, editing, and incorporating empirical feedback. While claude-opus-4.6 exhibits strong long-horizon optimization capabilities, most frontier models, including several proprietary ones, either terminate prematurely or exhaust their budgets with minimal progress. These results underscore the importance of time awareness and persistent iteration in autonomous agents. We open-source the full benchmark, evaluation harness, and task artifacts, to accelerate research toward truly capable long-horizon agents.

Zhangchen Xu, Junda Chen, Yue Huang +16

University of Washington · Bake AI · UCSD +11

Long-Horizon Agent EvaluationAI Agent BenchmarksLong-Horizon Agent TasksAutonomous Research Agents
AskPDF
Jun 3, 2026

Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents

Reinforcement learning with verifiable rewards improves reasoning and tool use, yet long-horizon language agents still learn unsupported evidence chains, belief drift, and shortcut actions that satisfy terminal checks. Existing process rewards are mostly correlational: they reward retrieval-, reflection-, or verification-like steps without estimating whether the step contributes to final verified success under a specified intervention. We propose CVT-RL, a constrained policy-gradient algorithm with dense verifiable rewards, intervention-validity gating, and a policy-conditioned counterfactual contribution (PCCC) estimator. Deletion, semantic substitution, evidence substitution, and tool-output perturbation define separate controlled interventions; continuations are sampled from a frozen reference policy, and a selection-adjusted doubly robust estimator augments the advantage. Belief control uses only prefix-observable labels, while an augmented Lagrangian constrains unsupported claims, skipped verification, tool tampering, and unsafe calls. On long-context QA, ALFWorld, ScienceWorld, and web/tool tasks, CVT-RL improves average task success from 71.8% for compute-matched non-causal RL and 75.4% for an information-matched counterfactual-process baseline to 78.9%, improves evidence F1 from 78.9 to 82.8 over the information-matched baseline, and reduces measured hacking from 7.2% to 3.9%. Independent human audit estimates 4.6% hacking for CVT-RL versus 8.1% for the information-matched baseline, and adaptive detector-evasion attacks raise hacking only to 7.1%. Stratified bootstrap and mixed-effects tests give p<0.01 after Holm correction for all primary metrics. Carefully scoped counterfactual credit, paired with validity gating, diagnostics, and verifiable constraints, provides a reproducible route toward more reliable long-horizon RL for language agents.

Renwei Meng

Ahu

Credit Assignment in RLLong-Horizon Agent EvaluationConstrained RLReinforcement Learning with Verifiable RewardsCounterfactual LearningRL for LLM Agents
AskPDF
Jun 3, 2026

LifeSide: Benchmarking Agents as Lifelong Digital Companions

Lifelong digital companions must integrate cross-session cues, continually update their understanding of users, and adapt to shifting privacy boundaries. Existing evaluations fail to capture this, testing memory recall and short-term empathy in isolation. To bridge this gap, we introduce \benchmark, a benchmark centered on multi-session Memory-Emotion-Environment loops. By modeling users as persistent worlds with layered profiles and event trajectories, \benchmark uses multi-agent simulation to project environmental dynamics into dialogue, preserving the critical gap between latent thoughts and observable expressions. Evaluating 2,000 personas and 111K tasks across memory tracking, user understanding, privacy control, and emotional companionship, our experiment results reveal a stark reality: even models that saturate current memory benchmarks fail to sustain accurate user understanding and true companionship over long horizons.

Yuqian Wu, Zhijie Deng, Wei Chen +8

Hong Kong University of Science and Technology (Guangzhou) · Tencent · Hong Kong University of Science and Technology

Long-Horizon Agent EvaluationAI CompanionsAI Agent BenchmarksMemory Agent BenchmarksLong-Term Conversational Memory
AskPDF
Jun 1, 2026

ClinEnv: An Interactive Multi-Stage Long Horizon EHR Environment for Agents

Clinical practice is not the selection of an answer from enumerated options: a physician gathers heterogeneous information incrementally and commits to sequential, irreversible decisions under uncertainty. Static benchmarks cannot probe and existing interactive medical benchmarks each compromise on at least one of them. We present ClinEnv, an interactive benchmark that evaluates LLMs as attending physicians over real inpatient admissions under a paradigm we term Longitudinal Inpatient Simulation. Each case is automatically constructed into an ordered sequence of decision stages; at every stage the model must actively query four specialized agents before committing to medications, procedures, and diagnoses. ClinEnv scores both what the model decides, through deterministic ontology-grounded matching, and how it gathers information. Across seven models, the strongest reaches only 0.31 decision F1, and outcome quality is sharply decoupled from process quality. Difficulty concentrates in management decisions and later stages, where models recover discharge diagnoses far more reliably than management actions (0.51 vs. 0.17 F1) and continue to issue redundant queries as cases progress. ClinEnv makes this information-acquisition gap, invisible to outcome-only evaluation, directly measurable.

Yuxing Lu, Yushuhong Lin, Wenqi Shi +4

1Georgia Institute of Technology · 2Peking University · University of Texas Southwestern Medical Center +1

Long-Horizon Agent EvaluationClinical Decision-MakingAI Agent BenchmarksActive Information Gathering
AskPDF
Jun 1, 2026

RescueBench: Can Embodied Agents Save Lives in the Wild ?

Search-and-rescue (SAR) requires embodied agents to explore unfamiliar environments under multimodal uncertainty, perform multi-stage interactions, and retrieve spatial memory over long horizons. Existing benchmarks typically evaluate these capabilities in isolation, leaving unclear how failures compound when they must be composed in realistic workflows. We introduce RescueBench, a photo-realistic diagnostic benchmark that instantiates SAR as a four-stage pipeline: multimodal exploration, target rescue, memory-guided return, and final handoff. By combining sequential task composition with stage-level evaluation, RescueBench enables analysis of how exploration and memory failures propagate through embodied rescue workflows. It contains five progressive difficulty levels that vary in environmental complexity, clue ambiguity, and spatial hierarchy, along with an automatic episode generation and annotation pipeline for scalable evaluation and training. We evaluate seven baselines, an oracle reference, and human players, showing that no baselines complete the full task at the greatest difficulty. Stage-level diagnosis identifies autonomous exploration as the dominant failure mode and spatial memory as a second, independent bottleneck, suggesting that these limitations are not resolved by current topological visual-language navigation or map-based methods. Code is available in https://github.com/wukui-muc/RescueBench

Kui Wu, Beiyu Guo, Hao Chen +6

1Beihang University · 5ATEC2025 Challenge Committee · 4City University of Macau +2

Agent MemoryEmbodied NavigationLong-Horizon Agent EvaluationRoboticsAI Agent BenchmarksLong-Term Memory for Embodied Agents+1
AskPDF
May 30, 2026

Momento: Evaluating Persistent Memory and Reasoning with Multi-Session Agentic Conversations

Recent advances in agentic AI have enabled agents to complete complex tasks through tool use, reasoning, and multi-step planning. Yet existing benchmarks evaluate agents within a single session, ignoring past actions, stated preferences, and prior decisions that agents must integrate to fulfill personalized user goals. We introduce Momento, a benchmark for persistent agentic task completion in multi-session service environments, requiring agents to take consequential, tool-mediated actions while resolving temporal dependencies and evolving user goals across sessions. Experimental results reveal that current agents fail primarily through misestimation of user state, treating prior session history as a reliable proxy for current context rather than stale information requiring re-validation, highlighting a substantial gap between current agent capabilities and realistic long-horizon human-agent interaction.

Adril Putra Merin, David Anugraha, Ayu Purwarianti +1

Institut Teknologi Bandung · Stanford University · Capital One

Agent MemoryTemporal ReasoningLong-Horizon Agent EvaluationAI Agent BenchmarksMemory Agent BenchmarksLong-Horizon Agent Tasks
AskPDF
May 29, 2026

AgentOdyssey: Open-Ended Long-Horizon Text Game Generation for Test-Time Continual Learning Agents

For agents to learn continuously from interaction with the world at test time, they must be able to explore effectively, acquire new world knowledge and skills, retain relevant episodic experiences, and plan over long horizons. To evaluate these key abilities of test-time continual learning agents, we introduce AgentOdyssey, a novel evaluation framework that procedurally generates open-ended text games with rich entities, world dynamics, and long-horizon tasks. Critically, AgentOdyssey goes beyond the conventional machine learning assumption that learning does not occur at test time by placing agents in a continuous, long-horizon setting that interleaves learning and inference throughout deployment. We further propose a multifaceted evaluation methodology that measures not only game progress but also offers diagnostic tests on world knowledge acquisition, episodic memory, object and action exploration, action diversity, and model cost. We evaluate diverse agent paradigms across six generated games. Our experimental results reveal critical limits in agents' key abilities, as well as factors that influence their meaningful horizon. Although performance scales with stronger base models, even the top agent remains far below human performance on a demanding long-horizon game, leaving substantial headroom for improvement. Additionally, a focused study of agent test-time training finds that, although short-term memory is beneficial, agents trained with supervised fine-tuning still suffer from catastrophic forgetting, suggesting that effective parametric continual learning remains an open challenge for test-time agents.

Zheyuan Zhang, Zehao Wen, Alvin Zhang +4

Johns Hopkins University

Continual Test-Time AdaptationLong-Horizon Agent EvaluationAI Agent Benchmarks
AskPDF
May 29, 2026

MineExplorer: Evaluating Open-World Exploration of MLLM Agents in Minecraft

Multimodal large language models (MLLMs) have shown strong capabilities in perception, reasoning, and action generation. However, their ability to sustain exploration in dynamic open worlds remains unclear. Existing embodied and game-based benchmarks often compress interaction into short-horizon tasks or entangle success with domain-specific game mechanics. In this paper, we introduce MineExplorer benchmark for evaluating open-world exploration capabilities of MLLM agents in Minecraft. We first filter atomic tasks whose solutions rely heavily on Minecraft-specific knowledge to better reflect general open-world reasoning. Then we organize the benchmark around a ReAct-style capability formulation and compose atomic tasks into implicit multi-hop tasks. To further construct reliable instances, MineExplorer uses a multi-agent synthesis workflow that jointly designs task graphs, sandbox scenes, and rule-based milestone evaluators. Human evaluation shows that the multi-agent synthesis workflow produces significantly more reliable instances than a single-agent baseline. Experiments with advanced MLLM agents show that open-world exploration remains challenging, as strong models can handle many single-hop tasks but degrade sharply when hidden prerequisites must be coordinated over longer trajectories. Further analysis finds that task difficulty tracks agent completion, and larger models or thinking modes do not consistently translate into better performance. Code and dataset are available at https://github.com/Jometeorie/MineExplorer.

Tianjie Ju, Yueqing Sun, Zheng Wu +7

School of Computer Science, Shanghai Jiao Tong University · 2Meituan

Long-Horizon Agent EvaluationAI Agent BenchmarksMultimodal Agents
AskPDF
May 28, 2026

LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis

Real-world data analysis is inherently iterative, yet existing benchmarks mostly evaluate isolated or short interactive tasks, leaving agents' ability to track evolving analytical context over long horizons untested. We introduce LongDS, a benchmark for long-horizon, multi-turn data analysis where agents must maintain, update, restore, and compose evolving analytical states. LongDS comprises 68 tasks constructed from real-world Kaggle notebooks, spanning 2,225 turns across six domains including Geoscience, Business, and Education. Tasks are designed around state-evolution patterns (e.g., counterfactual perturbation, rollback, multi-state composition), with an average dependency span of 11.3 turns. Evaluating five state-of-the-art models, we find that the best model reaches only 48.45% average accuracy, performance drops nearly 47 points from early to late turns, and long-horizon errors account for 52%--69% of failures. Further analysis shows that additional agent steps do not necessarily improve performance, suggesting that the key bottleneck is maintaining a correct analytical state rather than increasing interaction budget. We release LongDS to support research on reliable long-horizon agentic data analysis. Code and data will be released at https://github.com/zjunlp/DataMind.

Kewei Xu, Xiaoben Lu, Shuofei Qiao +4

Zhejiang University · Zhejiang University - Ant Group Joint Laboratory of Knowledge Graph · Ant Group

Long-Horizon Agent EvaluationLong-Horizon Agent TasksAgent BenchmarksData Analysis Agents
AskPDF
May 28, 2026

Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents

While GUI agents have advanced rapidly, they often lack the robustness to recover from their own errors, hindering real-world deployment. To bridge this gap at both the evaluation and data levels, we introduce GUI-RobustEval and propose Robustness-driven Trajectory Synthesis. GUI-RobustEval contains 1,2161,216 executable test cases that systematically measure error recovery capabilities across a broad and realistic spectrum of error modes. At the data level, RoTS is a scalable synthesis framework that creates 800k800k high-quality data via a tree-based pipeline that proactively discovers diverse error modes and synthesizes corresponding recovery steps. Our two models, RoTS-7B and RoTS-32B, fine-tuned on our dataset, both demonstrate significant gains on GUI-RobustEval and traditional GUI benchmarks. Notably, RoTS-32B achieves state-of-the-art performance on OSWorld, with a 47.4%47.4\% success rate and a 33.8%33.8\% All-Pass@4 score, suggesting that improved long-horizon error recovery ability contributes to both robustness and overall performance. Our code is available at https://github.com/AlibabaResearch/RoTS.

Tianpeng Bu, Xin Liu, Qihua Chen +7

Alibaba Cloud Computing.

GUI AgentsLong-Horizon Agent EvaluationLLM Agent Trajectory SynthesisGUI Agent Benchmarks
AskPDF
May 28, 2026

WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction

Multimodal large language models are increasingly deployed as long-horizon agents, where memory must do more than recall: it must track an evolving world, revise what has gone stale, and surface the right evidence at decision time. Existing benchmarks measure recall over static dialogue, collapse memory into a single end-of-task accuracy, and reduce visual observations to captions, leaving us unable to localize failures to writing, maintenance, retrieval, or use. The rise of agent harnesses that author their own memory sharpens this gap, since we have no principled way to compare hand-designed pipelines with self-managing alternatives. To close these gaps, we formulate multimodal agent memory as an Action-World Interaction Loop with an observable four-stage lifecycle, and instantiate it in WorldMemArena: 400 multi-session multimodal tasks spanning Lifelong Evolution (evolving personal and task states) and Agentic Execution (memory from real observations, actions, and feedback), annotated with gold memory points, updates, distractors, and evidence chains for stage-level diagnosis. This enables the first head-to-head comparison of long-context, manually designed (RAG and external memory systems), and harness-based memory agents. Results show that: (1) better memory writing and storage do not guarantee better performance; (2) multimodal memory still struggles to fully use visual evidence; (3) systems are unstable across domains and degrade on realistic agentic trajectories; and (4) harness memory is more flexible but remains costly and less reliable.

Chengzhi Liu, Yuzhe Yang, Sophia Xiao Pu +14

University of California, Santa Barbara · Johns Hopkins University · Carnegie Mellon University +3

Multimodal Agent EvaluationMultimodal MemoryLLM Agent MemoryAI Agent EvaluationLong-Horizon Agent EvaluationAI Agent Benchmarks+1
AskPDF
May 28, 2026

DynSess: Dynamic Session-Level Evaluation and Optimization Framework for Role-Playing Agents

Role-playing with large language models is fundamentally a session-level task, requiring agents to sustain character identity and interaction quality across extended multi-turn conversations. Yet existing evaluation and optimization methods remain largely turn-level, failing to capture long-horizon quality. We propose DynSess, a unified session-level framework for role-playing agents. DynSess-Eval scores complete dialogue sessions via rubrics targeting long-horizon behaviors. Leveraging its session-level rewards, we construct high-quality training trajectories through multi-turn lookahead search and train DynSess-Character with two complementary variants: DSPO (off-policy) and GSRPO (on-policy). Experiments show that DynSess-Eval aligns with human judgments substantially better than prior evaluators, and blind human evaluation further shows that DynSess-Character matches the strongest character model despite using substantially fewer parameters, while maintaining strong role consistency and interactive ability. Our dataset and code will be released to facilitate future research.

Rongsheng Zhang, Jiji Tang, Junnan Ren +6

1Zhejiang University · 2Fuxi AI Lab, NetEase Inc. · 3Xiamen University

Large Language Model-Based Role-Play SimulationLong-Horizon Agent EvaluationMulti-Turn Dialogue ManagementMulti-Turn Dialogue EvaluationLLM Agent TrainingConversational Agents+2
AskPDF
May 27, 2026

Verifiable Benchmarking of Long-Horizon Spatial Biology

AI agents are increasingly useful for biological data analysis, but existing benchmarks mostly test broad biological knowledge, executable workflows, or localized analysis steps rather than end-to-end scientific reasoning over spatial measurements. We introduce SpatialBench-Long, a benchmark for long-horizon spatial biology in which agents must recover biological claims from raw or near-raw data and calibrated experimental context without prescribed methods. SpatialBench-Long contains 24 evaluations across primary pancreatic ductal adenocarcinoma (PDAC), engineered glioblastoma organoids and in vivo tumors, Cas9 lineage-traced lung adenocarcinoma, and mouse optic nerve aging/intervention systems, spanning CosMx, Visium, Xenium, multiplexed error-robust fluorescence in situ hybridization (MERFISH), single-cell RNA sequencing (scRNA-seq), Slide-seq, Slide-tags, histology, and lineage-recording data. Candidate claims are hardened through reproduction, independent scientist review, and trajectory inspection. Final answers are graded deterministically over controlled vocabularies and symbols with companion rubrics capturing progress through key analysis chokepoints. Across the SpatialBench-Long benchmark, three model-harness pairs tie at 8/72 runs (11.1%): Gemini 3.5 Flash / Pi terminal coding harness, GPT-5.5 / Pi, and GPT-5.5 / OpenAI Codex. SpatialBench-Long tests whether agents can move beyond executing procedural analysis to deriving accurate scientific conclusions from complex spatial measurements.

Ian Diks, Harihara Muralidharan, Tim Proctor +1

LatchBio, San Francisco, CA, USA

Spatial TranscriptomicsBenchmark DesignLong-Horizon Agent EvaluationAI Agent Benchmarks
AskPDF
May 27, 2026

VibeSearchBench: Benchmarking Long-horizon Proactive Search in the Wild

LLM-based agents score well on search benchmarks, yet real users consistently find results unsatisfying, revealing a persistent evaluation-experience gap. We attribute this gap to existing benchmarks' reliance on over-specified queries, single-turn interactions, and fixed-schema evaluation, none of which reflect real search behavior where users and agents collaboratively refine vague intent through multi-turn dialogue. We term this paradigm VibeSearch and introduce VibeSearchBench, a benchmark comprising 200 manually curated bilingual (Chinese and English) tasks across 20 domains, split into VibeSearch-Pro (professional) and VibeSearch-Daily (daily-life) subsets. Each task pairs a user persona with a schema-free ground-truth knowledge graph, and is evaluated through a progressive-disclosure user simulator and a graph-matching evaluation framework. We benchmark seven frontier models under both the ReAct framework and the OpenClaw agent harness. Results show that all models remain substantially inadequate for VibeSearch (best F1: 30.30), highlighting the need for fundamental advances in long-context reasoning, proactive intent elicitation, and structured knowledge construction.

Xiaohongshu Inc

Xiaohongshu Dots Studio & Unipat AI

Agentic SearchLong-Horizon Agent EvaluationConversational SearchAI Agent Benchmarks
AskPDF
May 26, 2026

AndroidDaily: A Verifiable Benchmark for Mobile GUI Agents on Real-World Closed-Source Applications

The rapid development of GUI foundation models and mobile GUI agents has spurred numerous evaluation benchmarks, yet most rely on simulated environments or open-source applications, leaving real-world closed-source applications largely unevaluated. The core difficulty is that closed-source applications do not expose internal states, making traditional automatic verification inapplicable. To bridge this gap, we introduce AndroidDaily, a large-scale benchmark comprising 350 realistic daily-use tasks across 94 high-frequency Android applications spanning transportation, shopping, local services, entertainment, content creation, social media, and everyday utilities. To enable automatic and verifiable assessment in these opaque environments, we propose Guideline-grounded Reviewer for Automatic Diagnostic Evaluation (GRADE), a process-aware evaluator built on a three-tiered system of observable external guidelines: operational obligations, output quality, and negative constraints. GRADE tracks the agent's visual trajectory against these criteria and produces step-level diagnostic judgments, turning long-horizon, open-ended mobile interactions into verifiable evaluation without relying on hidden internal states. Experiments show that GRADE achieves 87.37% agreement with human evaluators. The strongest model reaches a 62.0% success rate on AndroidDaily, highlighting a substantial gap between current reasoning capabilities and practical execution in realistic mobile workflows.

Yifan Sui, Xin Huang, Hongbing Li +14

1Beijing University of Posts and Telecommunications · 2StepFun · 3Waseda University

Automated EvaluationComputer-Use Agent BenchmarksMobile GUI AutomationGUI AgentsLong-Horizon Agent EvaluationAI Agent Benchmarks+2
AskPDF
May 26, 2026

Benchmarks are Not Enough: RAMP for Runtime Assessing of Agentic Models in Production Systems

LLM agents are rapidly evolving from coding assistants into autonomous software engineering systems. However, existing evaluation methodologies remain largely centered on static, isolated, and short-horizon benchmarks that fail to capture the dynamic complexity of real-world production workflows. As a result, benchmark performance may poorly reflect practical capability under realistic runtime environments involving long execution chains, tool interactions, dependency management, and iterative feedback loops. We thus present RAMP, a production-grounded infrastructure for assessing long-horizon software engineering agents. Built upon the YatCC integrated platform, RAMP provides a unified runtime assessment architecture through standardized orchestration and execution interfaces. RAMP introduces realistic compiler-construction workloads with serial dependencies and complex toolchain interactions, together with a staged recovery mechanism for analyzing execution behavior under partial workflow failure. The framework further incorporates utility-oriented multi-dimensional metrics that jointly evaluate outcome quality and process efficiency. We conduct runtime assessments across 15 mainstream models and observe substantial capability degradation that remains largely invisible to conventional isolated benchmarks. Task completion rates progressively collapse across serial workflows, dropping from 100% in the initial stage to only 20% in the final stage, while none of the evaluated models successfully completes the entire pipeline. Runtime analysis reveals systematic failure propagation and significant resource inefficiencies, with computational costs differing by up to three orders of magnitude among comparable models. These findings suggest RAMP advances agentic model evaluation toward continuous, runtime-observable, and production-grounded assessment.

Yipeng Ouyang, Xin Huang, Bingjie Liu +3

School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China

Software Engineering AgentsLLM Agent EvaluationLong-Horizon Agent EvaluationAI Agent BenchmarksAI Agent Monitoring
AskPDF
May 26, 2026

VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactions

Large language models (LLMs) have evolved into interactive agents that collaborate with users in real-world tasks. Effective collaboration in such settings increasingly depends on understanding the user beyond what is explicitly stated, as user intent is often reflected in fragmented daily interactions and requires both personalized modeling and proactive interaction. However, existing agent benchmarks primarily evaluate reasoning and tool use, largely overlooking the challenges of inferring and leveraging user preferences in realistic scenarios. To address this gap, we introduce VitaBench 2.0, a benchmark for evaluating personalized and proactive agent behavior in long-term user interactions. In VitaBench 2.0, tasks are organized as temporally ordered sequences for individual users, where preferences are embedded in fragmented and heterogeneous interactions. Successful completion of tasks requires the agent to continuously extract, utilize, and update user preferences from these interactions. We further evaluate proactiveness through tasks that require agents to recognize missing information and actively acquire it from users or environments before making decisions. To support systematic analysis, we provide an extensible memory interface that enables controlled comparison across different memory architectures. We benchmark a diverse set of frontier proprietary and open-source LLMs. Results show that real-world personalization remains highly challenging even for state-of-the-art models, revealing a substantial gap between current capabilities and practical requirements. Extensive analysis further reveals the failure modes and capability bottlenecks of current agents in real-world personalized decision-making, providing insights for future model improvements.

Yuxin Chen, Yi Zhang, Zhengzhou Cai +11

1National University of Singapore · 2Meituan · University of Science and Technology of China +2

LLM Agent MemoryAI Agent EvaluationLong-Horizon Agent EvaluationAI Agent BenchmarksUser Preference Modeling
AskPDF
May 26, 2026

Persistent AI Agents in Academic Research: A Single-Investigator Implementation Case Study

Background: Large language models are typically evaluated as models, benchmarks, or short conversational episodes. Less is known about what happens when an agent is embedded persistently in a real academic research environment with durable memory, local files, external tools, scheduled routines, delegated roles, and explicit safety protocols. Methods: A structured self-observed implementation case study was conducted from January 31 to May 25, 2026. The unit of analysis was the persistent human-agent environment: researcher, agent runtime, memory layer, tools, repositories, scheduled jobs, specialized agent roles, and governance rules. Outcomes were organized using PARE-M (Persistent Agentic Research Environment Measurement), a measurement framework covering architecture, utilization, artifact production, resource use, reproducibility, and governance. Results: Recoverable main-agent telemetry contained 75,671 de-duplicated records across 96 active days, with 8,059 user-role and 23,710 assistant-role messages. The workspace included 502 memory-related files, 17 configured agent directories, and 57 skill files. Active system time was 579.7 hours (30-minute capped-gap estimate). Memory-derived records identified 482 output-proxy events and 889 failure, verification, correction, or protocol-proxy events. A strict May 2026 trajectory subset captured 627 model-completed events and 73.95 million recorded tokens, of which 82.9% were cache reads. Conclusions: The workflow was cache-dominant, suggesting that persistent agentic environments may shift the economic unit from cost per token to cost per completed artifact. Future evaluations should use artifact-level denominators, reproducible parsing rules, correction taxonomies, and independent coding of governance events.

Anas H. Alzahrani

Department of Preventive Medicine and Public Health, Faculty of Medicine, King Abdulaziz University, Jeddah, Saudi Arabia

Long-Horizon Agent EvaluationAI Agent GovernanceAutonomous Research Agents
AskPDF
May 25, 2026

Anchor: Mitigating Artifact Drift in Agent Benchmark Generation

AI agents are beginning to complete valuable, long-horizon business operations tasks, but training and evaluation environments for enterprise work still struggle to balance realism, verifiability, and scale. Environment and task creation frequently suffers from a failure mode we call artifact drift: when instructions, environments, oracles, and verifiers are created by loosely coupled processes, they frequently disagree on what a task requires, producing environments that are unsolvable, reward-hackable, or inconsistent. We introduce Anchor, a task-generation pipeline that formalizes domain experts' specifications of business workflows into constraint optimization programs. From a single parametric specification, the pipeline jointly produces a natural-language instruction, environment configuration, solver-certified ground-truth solution, and state-based verifier. With Anchor, altering parameters yields new tasks with controlled difficulty and known optimal solutions, producing harness-agnostic environments whose rewards depend solely on end-state business correctness. We apply Anchor to produce ERP-Bench: a benchmark of 300 long-horizon tasks spanning procurement and manufacturing workflows in a production-grade ERP system. We find that generation parameters predict realized difficulty, and that frontier models satisfy explicit task constraints in 26.1% of trials but reach a fully optimal solution in only 17.4% of trials. Overall, we show that Anchor and ERP-Bench offer a concrete recipe for building auditable evaluation environments for economically valuable agent work. We release the task generator and ERP-Bench dataset at erpbench.ai

Maksim Ivanov, Abhijay Rana

Agentic Labs New York, United States

Computer-Use Agent BenchmarksLong-Horizon Agent EvaluationAI Agent BenchmarksSynthetic Benchmark Generation
AskPDF
May 22, 2026

EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions

Coding agents are increasingly used as iterative development partners, but most benchmarks still evaluate one specification followed by one final assessment. This leaves out a basic question: can an agent keep its own codebase working as requirements change? We introduce EvoCode-Bench, a benchmark of 26 stateful coding tasks and 227 evaluated rounds. Each task preserves the agent's workspace for 5-15 rounds, states requirements through observable behavior, and uses cumulative executable tests to check new requirements and still-active prior ones. We evaluate 13 coding agents with two metrics: MT@4, a four-attempt fail-stop multi-round score, and SR, a single-round score from a reference-completed prior state. For most agents, SR exceeds MT@4 by 22-40 points. The gap also changes rankings: the highest-SR agent (78.9) ranks only third in persistent execution (44.0 MT@4). Even the strongest agents achieve only about 50% success on multi-turn metrics, and aggregate pass rate drops below half of round-1 performance by round 5. Failure analysis shows tier-dependent behavior: weaker agents fail early, while stronger agents survive long enough to expose specification-tracking and regression failures. We release the benchmark data and Harbor multi-turn infrastructure.

Haiyang Shen, Xuanzhong Chen, Wendong Xu +3

1UniPat AI · 2Peking University · 3Tsinghua University +1

AI Coding AgentsSoftware Engineering BenchmarksLong-Horizon Agent EvaluationAI Agent Benchmarks
AskPDF
May 22, 2026

Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents

Long-horizon language agents can make many plausible local tool calls yet fail to persist until a requested count is actually complete. We study this gap as Quantitative Goal Persistence (QGP): whether an agent keeps working until an external verifier confirms enough distinct valid items. PushBench turns this into a benchmark for repository-artifact collection and verifier-backed work units, so repeated work, duplicate submissions, false completion, and progress drift are measured directly rather than hidden behind a final success flag. In matched controller comparisons, a state-tracking retrieval controller reaches 69-78% success while eliminating duplicate submissions, and a backlog-tracking work-unit controller reaches 25-50% success in settings where standard and completion-gated controllers complete no task instances. Black-box frontier-agent evaluations with Claude Code (Sonnet 4.6) and Codex CLI (gpt-5.4) solve many 50-artifact tasks but drop to 3 out of 9 successes per condition at 100 artifacts. The results show that quantitative goals stress a different reliability requirement from local task competence: agents must maintain verified progress and stop only when the requested work is complete.

Yuandao Cai, Yuzhang Zhu, Liyou Gao +2

Independent Researcher · Xidian University

Long-Horizon PlanningLong-Horizon Agent EvaluationLLM Agent Reliability
AskPDF
May 21, 2026

TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks

We introduce TerminalWorld, a scalable data engine that automatically reverse-engineers high-fidelity evaluation tasks from "in-the-wild" terminal recordings. Processing 80,870 terminal recordings, the engine yields a full benchmark of 1,530 validated tasks, spanning 18 real-world categories, ranging from short everyday operations to workflows exceeding 50 steps, and covering 1,280 unique commands. From these, we curate a Verified subset of 200 representative, manually reviewed tasks. Comprehensive benchmarking on TerminalWorld-Verified across eight frontier models and six agents reveals that current systems still struggle with authentic terminal workflows, achieving a maximum pass rate of only 62.5%. Moreover, TerminalWorld captures real-world terminal capabilities distinct from existing expert-curated benchmarks (e.g., Terminal-Bench), with only a weak correlation to their scores (Pearson r=0.20). The automated engine makes TerminalWorld authentic and scalable by construction, enabling it to evaluate agents in real-world terminal environments as developer practices evolve. Data and code are available at https://github.com/EuniAI/TerminalWorld.

Zhaoyang Chu, Jiarui Hu, Xingyu Jiang +8

University College London · Nanjing University · Tencent

Computer-Use Agent BenchmarksTerminal-Agent BenchmarksTerminal AgentsLLM Agent EvaluationLong-Horizon Agent EvaluationAI Agent Benchmarks+2
AskPDF
May 20, 2026

SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents

As long-horizon coding agents produce more code than any developer can review, oversight collapses onto a single surface: the automated test suite. Reward hacking naturally arises in this setup, as the agent optimizes for passing tests while deviating from the users true goal. We study this reward hacking phenomenon by decompose software engineering tasks into three parts: (i) a natural language description of the specification (ii) visible validation tests that exercise specified features in isolation, and (iii) held-out tests that compose those same features to simulate real-world usage. Based on the specification and the visible validation test suites, a genuine agent would be able to generate a solution that can also pass all of the held-out tests. Therefore we use the gap in pass rates on these two suites to quantify reward hacking. Based on this methodology, we introduce SpecBench, a benchmark comprising 30 systems-level programming tasks ranging from short horizon tasks like building a JSON parser to ultra long horizon tasks like building an entire OS kernel from scratch. Large-scale experiments reveal a consistent pattern: while every frontier agent saturates the visible suite, reward hacking persists, with smaller models exhibiting larger gaps on holdout suites. The gap also scales sharply with task length: it grows by 28 percentage points for every tenfold increase in code size. Failures range from subtle feature isolation to deliberate exploits, including a 2,900-line hash-table "compiler" that memorizes test inputs. SpecBench offers a principled testbed for measuring whether coding agents build genuine working systems or merely game the test suites developers hand them.

Bingchen Zhao, Dhruv Srikanth, Yuxiang Wu +1

Weco AI

Reward HackingAI Coding AgentsSoftware Engineering BenchmarksLong-Horizon Agent Evaluation
AskPDF
May 19, 2026

Open-World Evaluations for Measuring Frontier AI Capabilities

Benchmark-based evaluation remains important for tracking frontier AI progress. But it can both overstate and understate deployed capability because it privileges tasks that can be precisely specified, automatically graded, easy to optimize for, and run with low budgets and short time horizons. We advocate for a complementary class of evaluations, which we term open-world evaluations: long-horizon, messy, real-world tasks assessed through small-sample qualitative analysis rather than benchmark-scale automation. In this paper we survey recent open-world evaluations, identify their strengths and limitations, and introduce CRUX (Collaborative Research for Updating AI eXpectations), a project for conducting such evaluations regularly. As a first instance, we task an AI agent with developing and publishing a simple iOS application to the Apple App Store. The agent completed the task with only a single avoidable manual intervention, suggesting that open-world evaluations can provide early warning of capabilities that may soon become widespread. We conclude with recommendations for designing and reporting open-world evals.

Sayash Kapoor, Peter Kirgis, Andrew Schwartz +15

Princeton University · Cornflower Labs · Meridian Labs +10

AI Agent EvaluationLong-Horizon Agent Evaluation
AskPDF