Search

4 results for “ALFWorld”

1. Research · Reinforcement learning

   AI agents complete more multistep tasks by reusing related experience

   BLPO: Beyond-Local Policy Optimization for Long-Horizon LLM Agents

   Across two simulated environments and two model sizes, the authors report higher task success when training reuses evidence about useful actions across tasks and training updates.

   ALFWorld: 94.01% In-Success (as reported)

   3 Nov 2026Beihang University, Alibaba Group

2. Research · Agents

   AI agents picking from a menu succeed more often with less training

   RLiD: Reinforcement Learning in the Decision Plane for Language Agents and LLMs

   On a household-task benchmark, training a model to choose directly among labelled actions reached 99.3% success in 12.7 GPU-hours, where token-level reinforcement learning reached 86.6% in 17.3.

   ALFWorld: 99.3% success (as reported)

   3 Nov 2026University of Illinois Urbana-Champaign, National University of Singapore, Zhejiang University

3. Research · Agents

   Structured backtracking beats reflection for long-horizon language agents

   Verify--Backtrack Loop: Structured Precondition Recovery for Long-Horizon Language Agents

   A training-free goal-stack controller lifts task success by up to 22 percentage points over ReAct across three benchmarks.

   ALFWorld: 87.31% task success rate (as reported)

   2 Nov 2026City University of Hong Kong, Huawei Noah's Ark Lab

4. Research · Agents

   RL method helps AI agents learn high-level task strategies from sparse rewards

   TalentRL: Strategy-Guided Reinforcement Learning for Hierarchical Skill Routing

   A hierarchical reinforcement learning approach separates strategy selection from skill execution to improve routing credit assignment.

   ALFWorld: 99.3% success rate (as reported)

   2 Nov 2026Microsoft AI