MedMemoryBench supplies a 2,000-session synthetic medical trajectory dataset and an evaluate-while-constructing streaming protocol to expose memory saturation and reasoning failures in current agent architectures for personalized healthcare.
super hub
WebArena: A Realistic Web Environment for Building Autonomous Agents
106 Pith papers cite this work. Polarity classification is still indexing.
106
Pith papers citing it
abstract
With advances in generative AI, there is now potential for autonomous agents to manage daily tasks via natural language commands. However, current agents are primarily created and tested in simplified synthetic environments, leading to a disconnect with real-world scenarios. In this paper, we build an environment for language-guided agents that is highly realistic and reproducible. Specifically, we focus on agents that perform tasks on the web, and create an environment with fully functional websites from four common domains: e-commerce, social forum discussions, collaborative software development, and content management. Our environment is enriched with tools (e.g., a map) and external knowledge bases (e.g., user manuals) to encourage human-like task-solving. Building upon our environment, we release a set of benchmark tasks focusing on evaluating the functional correctness of task completions. The tasks in our benchmark are diverse, long-horizon, and designed to emulate tasks that humans routinely perform on the internet. We experiment with several baseline agents, integrating recent techniques such as reasoning before acting. The results demonstrate that solving complex tasks is challenging: our best GPT-4-based agent only achieves an end-to-end task success rate of 14.41%, significantly lower than the human performance of 78.24%. These results highlight the need for further development of robust agents, that current state-of-the-art large language models are far from perfect performance in these real-life tasks, and that WebArena can be used to measure such progress.
hub tools
citation-role summary
dataset 1
citation-polarity summary
claims ledger
- abstract With advances in generative AI, there is now potential for autonomous agents to manage daily tasks via natural language commands. However, current agents are primarily created and tested in simplified synthetic environments, leading to a disconnect with real-world scenarios. In this paper, we build an environment for language-guided agents that is highly realistic and reproducible. Specifically, we focus on agents that perform tasks on the web, and create an environment with fully functional websites from four common domains: e-commerce, social forum discussions, collaborative software develop
authors
co-cited works
roles
dataset 1polarities
use dataset 1representative citing papers
Agent-BRACE improves LLM agent performance on long-horizon partially observable tasks by 5.3-14.5% through a decoupled belief state of verbalized atomic claims with certainty labels that keeps context length constant.
A new native-runtime benchmark reveals that current frontier AI agents succeed on at most 62 percent of realistic long-horizon CLI tasks.
WindowsWorld benchmark shows leading GUI agents achieve under 21% success on multi-application professional tasks, with failures especially on conditional judgment across three or more apps and inefficient execution.
AgentDojo introduces an extensible evaluation framework populated with realistic agent tasks and security test cases to measure prompt injection robustness in tool-using LLM agents.
OSWorld provides the first unified real-computer benchmark for open-ended multimodal agent tasks, exposing large performance gaps between humans and state-of-the-art LLM/VLM agents.
10.7% of passing SWE-agent trajectories are Lucky Passes with chaotic behaviors, and a quality score based on process references changes model rankings across eight backends.
SDP constructs a task-induced state space from raw text by having agents commit to and certify natural-language predicates as states, enabling structured planning and analysis in unstructured language environments.
BenchJack audits 10 AI agent benchmarks, synthesizes exploits achieving near-perfect scores without task completion, surfaces 219 flaws, and reduces hackable-task ratios to under 10% on four benchmarks via iterative patching.
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
Checkup2Action is a new multimodal dataset and benchmark for generating safe, prioritized action cards from real-world clinical check-up reports using large language models.
Mobius Injection exploits semantic closure in LLM agents to enable single-message AbO-DDoS attacks achieving up to 51x call amplification and 229x latency inflation.
ReVision reduces visual token usage by 46% on average in agent trajectories via a learned patch selector and improves success rates by 3% on three benchmarks, showing that history saturation stems from inefficient representations rather than lack of utility.
TacoMAS performs test-time co-evolution of agent capabilities and communication topology in LLM multi-agent systems via fast capability updates and slow meta-LLM topology edits, delivering 13.3% average gains over strong baselines on four benchmarks.
AgentForesight trains a 7B model to perform online auditing of multi-agent LLM trajectories, detecting early decisive errors and outperforming larger models on custom and external benchmarks.
LLM agents exhibit persistent attack-selection biases as fixed traits independent of success rates, with a bias momentum effect that resists steering and yields no performance gain.
A new evaluation protocol shows agent memory reliability degrades variably with added irrelevant sessions depending on agent, memory interface, and scale.
LLM agents reach only 50.6% accuracy on chemical cost estimation within 25% error even with tools, dropping with noise due to parsing, pack selection, and tool-use failures.
Weblica scales RL training for visual web agents by building thousands of reproducible environments through HTTP caching for stable replays and LLM synthesis from real sites, yielding an 8B model that beats similar open baselines on navigation benchmarks.
A VOI-based controller for dual inference budgets improves multi-hop QA performance by prioritizing search actions and selectively finalizing answers.
NeuroState-Bench is a human-calibrated benchmark with 144 tasks and 306 side-query probes showing that commitment integrity in LLM agent profiles diverges from task success, with 31 of 32 profiles changing rank under integrity evaluation.
DV-World is a benchmark of 260 tasks across spreadsheet manipulation, visual evolution, and interactive intent alignment that shows state-of-the-art AI models achieve less than 50% overall performance on real-world data visualization challenges.
ClawMark is a new benchmark for multi-turn multi-day multimodal coworker agents in stateful evolving services, with deterministic Python checkers showing frontier models achieve only 20% strict task success.
A new 7x4 taxonomy organizes agentic AI security threats by architectural layer and persistence timescale, revealing under-explored upper layers and missing defenses after surveying 116 papers.
citing papers explorer
-
MedMemoryBench: Benchmarking Agent Memory in Personalized Healthcare
MedMemoryBench supplies a 2,000-session synthetic medical trajectory dataset and an evaluate-while-constructing streaming protocol to expose memory saturation and reasoning failures in current agent architectures for personalized healthcare.
-
Agent-BRACE: Decoupling Beliefs from Actions in Long-Horizon Tasks via Verbalized State Uncertainty
Agent-BRACE improves LLM agent performance on long-horizon partially observable tasks by 5.3-14.5% through a decoupled belief state of verbalized atomic claims with certainty labels that keeps context length constant.
-
WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation
A new native-runtime benchmark reveals that current frontier AI agents succeed on at most 62 percent of realistic long-horizon CLI tasks.
-
WindowsWorld: A Process-Centric Benchmark of Autonomous GUI Agents in Professional Cross-Application Environments
WindowsWorld benchmark shows leading GUI agents achieve under 21% success on multi-application professional tasks, with failures especially on conditional judgment across three or more apps and inefficient execution.
-
AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
AgentDojo introduces an extensible evaluation framework populated with realistic agent tasks and security test cases to measure prompt injection robustness in tool-using LLM agents.
-
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
OSWorld provides the first unified real-computer benchmark for open-ended multimodal agent tasks, exposing large performance gaps between humans and state-of-the-art LLM/VLM agents.
-
AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation
10.7% of passing SWE-agent trajectories are Lucky Passes with chaotic behaviors, and a quality score based on process references changes model rankings across eight backends.
-
State-Centric Decision Process
SDP constructs a task-induced state space from raw text by having agents commit to and certify natural-language predicates as states, enabling structured planning and analysis in unstructured language environments.
-
Do Androids Dream of Breaking the Game? Systematically Auditing AI Agent Benchmarks with BenchJack
BenchJack audits 10 AI agent benchmarks, synthesizes exploits achieving near-perfect scores without task completion, surfaces 219 flaws, and reduces hackable-task ratios to under 10% on four benchmarks via iterative patching.
-
Covering Human Action Space for Computer Use: Data Synthesis and Benchmark
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
-
Checkup2Action: A Multimodal Clinical Check-up Report Dataset for Patient-Oriented Action Card Generation
Checkup2Action is a new multimodal dataset and benchmark for generating safe, prioritized action cards from real-world clinical check-up reports using large language models.
-
Can a Single Message Paralyze the AI Infrastructure? The Rise of AbO-DDoS Attacks through Targeted Mobius Injection
Mobius Injection exploits semantic closure in LLM agents to enable single-message AbO-DDoS attacks achieving up to 51x call amplification and 229x latency inflation.
-
ReVision: Scaling Computer-Use Agents via Temporal Visual Redundancy Reduction
ReVision reduces visual token usage by 46% on average in agent trajectories via a learned patch selector and improves success rates by 3% on three benchmarks, showing that history saturation stems from inefficient representations rather than lack of utility.
-
TacoMAS: Test-Time Co-Evolution of Topology and Capability in LLM-based Multi-Agent Systems
TacoMAS performs test-time co-evolution of agent capabilities and communication topology in LLM multi-agent systems via fast capability updates and slow meta-LLM topology edits, delivering 13.3% average gains over strong baselines on four benchmarks.
-
AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems
AgentForesight trains a 7B model to perform online auditing of multi-agent LLM trajectories, detecting early decisive errors and outperforming larger models on custom and external benchmarks.
-
CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios
LLM agents exhibit persistent attack-selection biases as fixed traits independent of success rates, with a bias momentum effect that resists steering and yields no performance gain.
-
When Stored Evidence Stops Being Usable: Scale-Conditioned Evaluation of Agent Memory
A new evaluation protocol shows agent memory reliability degrades variably with added irrelevant sessions depending on agent, memory interface, and scale.
-
Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning
LLM agents reach only 50.6% accuracy on chemical cost estimation within 25% error even with tools, dropping with noise due to parsing, pack selection, and tool-use failures.
-
Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
Weblica scales RL training for visual web agents by building thousands of reproducible environments through HTTP caching for stable replays and LLM synthesis from real sites, yielding an 8B model that beats similar open baselines on navigation benchmarks.
-
Inference-Time Budget Control for LLM Search Agents
A VOI-based controller for dual inference budgets improves multi-hop QA performance by prioritizing search actions and selectively finalizing answers.
-
NeuroState-Bench: A Human-Calibrated Benchmark for Commitment Integrity in LLM Agent Profiles
NeuroState-Bench is a human-calibrated benchmark with 144 tasks and 306 side-query probes showing that commitment integrity in LLM agent profiles diverges from task success, with 31 of 32 profiles changing rank under integrity evaluation.
-
DV-World: Benchmarking Data Visualization Agents in Real-World Scenarios
DV-World is a benchmark of 260 tasks across spreadsheet manipulation, visual evolution, and interactive intent alignment that shows state-of-the-art AI models achieve less than 50% overall performance on real-world data visualization challenges.
-
ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents
ClawMark is a new benchmark for multi-turn multi-day multimodal coworker agents in stateful evolving services, with deterministic Python checkers showing frontier models achieve only 20% strict task success.
-
A Systematic Survey of Security Threats and Defenses in LLM-Based AI Agents: A Layered Attack Surface Framework
A new 7x4 taxonomy organizes agentic AI security threats by architectural layer and persistence timescale, revealing under-explored upper layers and missing defenses after surveying 116 papers.
-
AJ-Bench: Benchmarking Agent-as-a-Judge for Environment-Aware Evaluation
AJ-Bench provides 155 tasks in three domains to evaluate environment-interacting agent judges, showing performance gains over LLM-as-a-Judge but exposing remaining verification challenges.
-
From Imitation to Discrimination: Progressive Curriculum Learning for Robust Web Navigation
A new 590k-instance dataset built with hard-negative mining and dual-agent verification, plus progressive SFT-to-ORPO-to-GRPO training, yields 58.7% step success on Mind2Web, beating GPT-4.5 and Claude-4.5.
-
ClawVM: Harness-Managed Virtual Memory for Stateful Tool-Using LLM Agents
ClawVM introduces a harness-managed virtual memory system for LLM agents that ensures deterministic residency and durability of state under token budgets by using typed pages and validated writeback.
-
EE-MCP: Self-Evolving MCP-GUI Agents via Automated Environment Generation and Experience Learning
A self-evolving MCP-GUI agent system with automated environment generation and an experience bank achieves up to 77.8% pass rates by matching distillation or experience augmentation to task type across three desktop applications.
-
HiL-Bench (Human-in-Loop Benchmark): Do Agents Know When to Ask for Help?
HiL-Bench shows frontier AI agents fail to ask for help on incomplete tasks, recovering only a fraction of full-information performance, but RL training on Ask-F1 reward improves judgment and transfers across domains.
-
SAGE: A Service Agent Graph-guided Evaluation Benchmark
SAGE is a new multi-agent benchmark that formalizes service SOPs as dynamic dialogue graphs to measure LLM agents on logical compliance and path coverage, uncovering an execution gap and empathy resilience across 27 models in 6 scenarios.
-
MolmoWeb: Open Visual Web Agent and Open Data for the Open Web
Open 4B and 8B visual web agents achieve state-of-the-art results on browser benchmarks by predicting actions from screenshots and instructions, outperforming similar open models and some closed larger-model agents, with full release of data and code planned.
-
WebSP-Eval: Evaluating Web Agents on Website Security and Privacy Tasks
WebSP-Eval shows that multimodal LLM-based web agents fail more than 45% of the time on security and privacy tasks involving stateful UI elements such as toggles and checkboxes.
-
GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis
GUIDE decomposes GUI agent evaluation into trajectory segmentation, subtask diagnosis, and overall summary to deliver higher accuracy and structured error reports than holistic baselines.
-
M-CARE: Standardized Clinical Case Reporting for AI Model Behavioral Disorders, with a 20-Case Atlas and Experimental Validation
M-CARE provides a medical-inspired reporting system for AI behavioral disorders, demonstrated through 20 cases and a validated experiment showing shell instructions overriding cooperative behavior across game domains.
-
$\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
τ²-bench provides a Dec-POMDP-based telecom domain with compositional task generation and a tool-constrained user simulator to measure agent performance drops in dual-control versus single-control settings.
-
$\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
τ-bench shows state-of-the-art agents like GPT-4o succeed on under 50% of tool-using, rule-following tasks and are inconsistent across repeated trials.
-
WAAA! Web Adversaries Against Agentic Browsers
Agentic browsers are vulnerable to 20 web and LLM attacks with 18 implemented, exposing five failure modes across four major LLM models that require redesign before safe deployment.
-
AcademiClaw: When Students Set Challenges for AI Agents
AcademiClaw is a new benchmark of 80 student-sourced academic tasks where the best frontier AI agents achieve only a 55% pass rate.
-
Beyond Chat and Clicks: GUI Agents for In-Situ Assistance via Live Interface Transformation
GUI agents can transform live web interfaces in real-time via DOM manipulations to deliver contextual assistance directly within the application.
-
RiskWebWorld: A Realistic Interactive Benchmark for GUI Agents in E-commerce Risk Management
RiskWebWorld is the first realistic interactive benchmark for GUI agents in e-commerce risk management, revealing a large gap between generalist and specialized models plus RL gains.
-
FrontierFinance: A Long-Horizon Computer-Use Benchmark of Real-World Financial Tasks
FrontierFinance benchmark shows human financial experts outperform state-of-the-art LLMs by achieving higher scores and more client-ready outputs on realistic long-horizon tasks.
-
How to Interpret Agent Behavior
ACT*ONOMY is a Grounded-Theory-derived hierarchical taxonomy and open repository that enables systematic comparison and characterization of autonomous agent behavior across trajectories.
-
MMSkills: Towards Multimodal Skills for General Visual Agents
MMSkills turns public interaction trajectories into compact multimodal skill packages that visual agents can consult at runtime to improve decision-making on benchmarks.
-
MMCL-Bench: Multimodal Context Learning from Visual Rules, Procedures, and Evidence
MMCL-Bench shows that even the strongest frontier multimodal models solve fewer than one-third of tasks requiring recovery and application of visual rules, procedures, and empirical patterns.
-
SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces
SkillSafetyBench shows that localized non-user attacks via skills and artifacts can consistently induce unsafe agent behavior across domains and model backends, independent of user intent.
-
Domain Restriction via Multi SAE Layer Transitions
Multi-layer SAE transitions capture domain-specific signatures that distinguish OOD texts in Gemma-2 models.
-
On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
FATE lets LLM agents self-evolve safer behaviors by generating and filtering repairs from their own failure trajectories using verifiers and Pareto optimization.
-
ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox
ComplexMCP benchmark shows current LLM agents achieve at most 60% success on interdependent tool tasks versus 90% for humans, due to tool retrieval saturation, over-confidence, and strategic defeatism.
-
Can Agent Benchmarks Support Their Scores? Evidence-Supported Bounds for Interactive-Agent Evaluation
Agent benchmarks can report evidence-supported score bounds instead of single misleading success rates by adding a layer that checks required artifacts for outcome verification.
-
TIDE-Bench: Task-Aware and Diagnostic Evaluation of Tool-Integrated Reasoning
TIDE-Bench is a new benchmark for tool-integrated reasoning that combines diverse tasks, multi-aspect metrics covering answer quality, process reliability, efficiency and cost, plus filtered challenging test sets.