Beyond Browsing: API-Based Web Agents
Yueqi Song , Frank F. Xu , Shuyan Zhou , Graham Neubig
- 🏛 Institutions
- CMU
- 📅 Date
- October 24, 2024
- 📑 Publisher
- Findings of ACL 2025
- 💻 Env
- Web
- 🔑 Keywords
TLDR
Studies what happens when web-agent tasks are solved through APIs instead of only through browsers. The paper proposes both API-only and hybrid agents, and shows that hybrid access to APIs plus browsing substantially outperforms browsing alone on WebArena.
Related papers (24)
- The Tool Illusion: Rethinking Tool Use in Web AgentsApril 3, 2026 · arXiv
- When Users Change Their Mind: Evaluating Interruptible Agents in Long-Horizon Web NavigationApril 1, 2026 · arXiv
- WebArena-Infinity: Generating Browser Environments with Verifiable Tasks at ScaleMarch 2026 · Blog Post
- AI Planning Framework for LLM-Based Web AgentsMarch 13, 2026 · arXiv
- HATS: Hardness-Aware Trajectory Synthesis for GUI AgentsMarch 12, 2026 · CVPR 2026
- WebWorld: A Large-Scale World Model for Web Agent TrainingFebruary 16, 2026 · arXiv
- OpAgent: Operator Agent for Web NavigationFebruary 14, 2026 · arXiv
- DynaWeb: Model-Based Reinforcement Learning of Web AgentsJanuary 29, 2026 · arXiv
- ColorBrowserAgent: Complex Long-Horizon Browser Agent with Adaptive Knowledge EvolutionJanuary 12, 2026 · arXiv
- WebOperator: Action-Aware Tree Search for Autonomous Agents in Web EnvironmentDecember 14, 2025 · arXiv
- WALT: Web Agents that Learn ToolsOctober 1, 2025 · ICLR 2026 (Poster)
- Go-Browse: Training Web Agents with Structured ExplorationJune 4, 2025 · ICLR 2026 (Poster)
- Inducing Programmatic Skills for Agentic TasksApril 9, 2025 · COLM 2025
- Large Language Models Can Self-Improve At Web Agent TasksMay 30, 2024 · arXiv
- SteP: Stacked LLM Policies for Web ActionsOctober 5, 2023 · COLM 2024
- WebArena: A Realistic Web Environment for Building Autonomous AgentsJuly 25, 2023 · ICLR 2024 (Poster)
- R-WoM: Retrieval-augmented World Model For Computer-use AgentsOctober 13, 2025 · ICLR 2026 (Poster)
- GUI Agents for Continual Game GenerationMay 27, 2026 · arXiv
- Odysseys: Benchmarking Web Agents on Realistic Long Horizon TasksApril 27, 2026 · arXiv
- WebForge: Breaking the Realism-Reproducibility-Scalability Trilemma in Browser Agent BenchmarkApril 13, 2026 · arXiv
- The Amazing Agent Race: Strong Tool Users, Weak NavigatorsApril 11, 2026 · arXiv
- Same Outcomes, Different Journeys: A Trace-Level Framework for Comparing Human and GUI-Agent Behavior in Production Search SystemsApril 9, 2026 · arXiv
- MolmoWeb: Open Visual Web Agent and Open Data for the Open WebApril 9, 2026 · arXiv
- ClawBench: Can AI Agents Complete Everyday Online Tasks?April 9, 2026 · arXiv