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
2 RSS
Relatedreinforcement-learning4webshop3scienceworld2agentic-rl1grpo1language-agents1llm-agents1oracle-continuations1ppo1proas1qwen1qwen2.51sample-efficiency1sero1world-models1
14 stories, newest first · one card per story
Stories tagged #alfworld
ProAS reshapes advantages for multi-turn agent reinforcement learning
On Qwen2.5 instruct models, reported peaks reach 94.3% on ALFWorld and 70.8% on WebShop.
Co-Evolving Acting and Forecasting for Language Agents
CoPE trains language agents to act and forecast future actions, improving benchmark scores and sample efficiency.
SERO: Learning from Executable Oracle Continuations for Long-Horizon Agentic Reinforcement Learning
SERO trains long-horizon agents using executable oracle continuations and reachability weighting on ALFWorld and WebShop.
Branching On-Policy Distillation
BranchOPD branches uncertain on-policy rollouts, beating vanilla OPD by 18.55 math and 13.8 ALFWorld points.
UniSkill: Learning Actor-Aligned Skill Proposals for an Evolving Policy
UniSkill lets one policy act and propose skill edits using contrastive feedback without extra rollouts.
Semantic Behavioral Watermarking: Paraphrase-Robust and Forgery-Resistant Provenance for LLM Agents
Semantic Behavioral Watermarking marks LLM-agent actions so provenance survives paraphrasing and resists forged trajectories.
ASCENT: Online Test-Time Training of Long-Horizon Agents via Self-Distillation of Verified Experience
ASCENT updates deployed LLM agents by self-distilling verified trajectories into LoRA weights at test time.
PivotOPD: Learning to Recover from Pivotal Mistakes in Multi-Turn Agents
PivotOPD trains agents to avoid and recover from early pivotal mistakes, lifting ALFWorld and SWE-Bench scores.
PivotOPD: Learning to Recover from Pivotal Mistakes in Multi-Turn Agents
PivotOPD trains multi-turn agents to avoid pivotal mistakes and recover within a few turns.
Do LLM Agents Execute the Plans They Declare? From Planning-Mode Declaration to Pattern-Specific Execution
Pattern-specific executors close LLM agents' plan-execution gap, lifting ALFWorld success from 0.48 to 0.92.
ControlScope: Workflow Revision and Reliability in LLM Agents
ControlScope finds only small, mixed gains from revising more of a running LLM-agent workflow.
Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents
Just-in-Time Memory curates raw agent trajectories at read time and beats write-time memory on three benchmarks.
RetireOPD: Self-Retiring On-Policy Distillation for Agentic Reinforcement Learning
RetireOPD adds self-retiring on-policy distillation to agentic RL, boosting Qwen2.5 agent success on ALFWorld by 14-19% and WebShop by 12-19%.