Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning
2025-01-22
· ICLR 2025 Poster ·
anchor
Findings
IC-477
Released LLMs achieve limited success rates as web agents on WebArena-Lite, with open-source models substantially below proprietary ones
IC-478
GPT-4 and GPT-4V achieve approximately 71-73% accuracy in judging whether a web agent trajectory successfully completes a task