Self-Evolving World Models for LLM Agent Planning
cs.AI, cs.CL
Submitted: 2026-06-29
Updated: 2026-09-01
Comments: Accepted at EMNLP 2026 Findings
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Scaling Agent Learning via Experience Synthesis
- Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
- DynaWeb: Model-Based Reinforcement Learning of Web Agents
- Model-Free Episodic Control
- Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents
- GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
- World Models
- Mastering Diverse Domains through World Models
- Imagine-then-Plan: Agent Learning from Adaptive Lookahead with World Models
- LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
- R-Zero: Self-Evolving Reasoning LLM from Zero Data
- MemGPT: Towards LLMs as Operating Systems
- Co-Evolving Agents: Learning from Failures as Hard Negatives
- Current Agents Fail to Leverage World Model as Tool for Foresight
- A Comprehensive Survey on World Models for Embodied AI
- From Word to World: Can Large Language Models be Implicit Text-based World Models?
- Self-Improving World Modelling with Latent Actions
- Reward Prediction with Factorized World States
- Guided Self-Evolving LLMs with Minimal Human Supervision
- Reinforcement World Model Learning for LLM-based Agents
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems