Steering Black-Box Large Language Model Agents with Pitfall Anticipation World Model

Introduces Pitfall Anticipation World Model (PAWM), which selectively intervenes to predict impending failures within a bounded future horizon, improving Claude Sonnet 4.5 performance on WebArena-Lite and ScienceWorld.

• Trace2World: Reconstructing App Worlds from Real Interaction Traces for Scalable GUI Agent Training
  October 22, 2026 · arXiv
• Agent World Model in the Wild: Computer-Use Agent Environment Synthesis
  October 19, 2026 · arXiv
• Beyond Next-Screen Prediction: The Decision Value of GUI World Models
  October 17, 2026 · arXiv
• Right Screen, Wrong Transition: World Models as Verifiers for GUI Agents
  October 8, 2026 · arXiv
• Harness Evolution Hits a Ceiling: When Weight Training Should Begin
  October 8, 2026 · arXiv
• Imagine to Act: High-Fidelity Data Synthesis via Image Editing World Model for Scalable GUI Agent Training
  October 5, 2026 · arXiv
• AutoGUIWorld: Image Generators as Visual World Models for GUI Agent
  October 1, 2026 · arXiv
• Auditing Web Agent Evaluation on WebArena-Lite: Human Review of Outcomes and Trajectories
  October 1, 2026 · NeurIPS 2026 Workshop on Who Verifies the Agents? Toward Reliable Agent Development