Self-Improving Agents with World Model Predictive Paths
We propose a test-time framework, inspired by Model Predictive Path Integral (MPPI) control, that combines multi-branch predictive analysis with experience-based adaptation.
ProofPaper ↗
Key points
- Large language model (LLM) agents are increasingly expected to interact with environments and improve their decisions during deployment.
- However, how anticipating future outcomes and drawing on accumulated interaction experience can jointly support decision-making remains underexplored.
- A dual-branch experience memory enables self-improvement: the planner branch refines candidate generation and action selection, while the world model branch improves next-observation prediction.
- Experiments on ALFWorld and ScienceWorld across three backbones demonstrate the strong overall performance of our method and the effectiveness of experience-guided predictive control.
Sources (1)
- [1]Self-Improving Agents with World Model Predictive PathsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 20, 09:46 PM
We propose a test-time framework, inspired by Model Predictive Path Integral (MPPI) control, that combines multi-branch predictive analysis with experience-based adaptation.
Large language model (LLM) agents are increasingly expected to interact with environments and improve their decisions during deployment.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 20, 2026Discover tools at runtime with AWS Agent Registry
- Oct 20, 2026Voxtral 3
- Oct 20, 2026SAPD: Learning from Agent Experience without Environment Interaction
- Oct 20, 2026Do Diffusion LLMs Scale? Testing Mercury 3 + Decide
- Oct 20, 2026OverSeer: Multi-Agent World Models Beyond What Agents See
- Oct 20, 2026Learning the 4D World through Point Trajectory Forecasting