WebArena

WebArena is a benchmark dataset used to evaluate the performance of large language model web agents by measuring their ability to execute structured tool actions based on web interactions.

Papers using WebArena (20)

Mobile-agent-v3.5: Multi-platform Fundamental GUI Agents2026Multi-Agent Transactive Memory2026Devil's Advocate: Anticipatory Reflection For LLM Agents2024 · 3 citesThe Compliance Trap: Diagnosing How AI Agents Consume Conflicting Memory2026Online Skill Learning for Web Agents via State-Grounded Dynamic Retrieval2026Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns2026Weasel: Out-of-Domain Generalization for Web Agents via Importance-Diversity Data Selection2026APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents2026Adarubric: Task-adaptive Rubrics For LLM Agent Evaluation2026Does The Way You Plan Matter? An Empirical Study of Planning Representations for LLM Web Agents2026Branch-and-Browse: Efficient and Controllable Web Exploration with Tree-Structured Reasoning and Action Memory2025WebUncertainty: Dual-Level Uncertainty Driven Planning and Reasoning For Autonomous Web Agent2026Agenther: Hindsight Experience Replay For LLM Agent Trajectory Relabeling2026Environment Maps: Structured Environmental Representations For Long-horizon Agents2026OpAgent: Operator Agent for Web Navigation2026Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates2026SkyRL-Agent: Efficient RL Training for Multi-turn LLM Agent2025WEBSERV: A Full-Stack and RL-Ready Web Environment for Training Web Agents at Scale2026Surfer 2: The Next Generation Of Cross-platform Computer Use Agents2025CoAct: A Global-Local Hierarchy for Autonomous Agent Collaboration2024 · 1 cites