WM-PPO values world-model predictions to improve sparse-reward agents
With a similar number of real steps, success rises from 63.5% to 87.5% on ALFWorld and from 39.3% to 52.9% on ScienceWorld.
Stories per day · 30 days
1 RSS
Relatedalfworld2grpo2reinforcement-learning2world-models2language-agents1llm-agents1ppo1sample-efficiency1webarena1webshop1
6 stories, newest first · one card per story
Stories tagged #scienceworld
PAWM steers black-box LLM agents by anticipating pitfalls
Authors report Claude Sonnet 4.5 gains of 24.3% on ScienceWorld and 7.7% on WebArena-Lite.
Co-Evolving Acting and Forecasting for Language Agents
CoPE trains language agents to act and forecast future actions, improving benchmark scores and sample efficiency.
Learning from Teacher Continuations at Student States
OLIVE updates a student on teacher continuations of student-generated prefixes, beating offline distillation under the same budget.
X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization
X-Tree tokenizes reusable agent skills and raises success on WebArena, ScienceWorld, and WebShop.
Cognitive Extensions for Dual-Process Language Agents: Memory and Self-Reflection in Interactive Environments
Researchers extend the SwiftSage dual-process agent with adaptive memory and self-reflection modules, improving scores in interactive environments.