Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL

arXiv:2607.16204 · cs.AI, cs.LG · Submitted 2026-05-07 · Read on arXiv

cs.AI, cs.LG

Submitted: 2026-05-07

Updated: 2026-10-07

Code: https://github.com/patronus-ai/mdlm_world_modeling

Terminology

• World Model — 3× in this paper · explained in 7 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Adversarial scenarios — 1× in this paper · explained in 1 episode
• Failure Modes — 1× in this paper · explained in 3 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• GRPO — 1× in this paper · explained in 6 episodes
• Mode Collapse — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes
• Steering — 1× in this paper · explained in 2 episodes
• Zero-Shot Transfer — 1× in this paper · explained in 5 episodes

Sources

• LFM2 Technical Report
• Large Language Models Hallucination: A Comprehensive Survey
• Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models
• AbideGym: Turning Static RL Worlds into Adaptive Challenges
• From Masks to Worlds: A Hitchhiker's Guide to World Models
• The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"
• LLaDA2.0: Scaling Up Diffusion Language Models to 100B
• LLaDA2.1: Speeding Up Text Diffusion via Token Editing
• Scaling Agent Learning via Experience Synthesis
• SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generation
• Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective
• SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
• Process Reinforcement through Implicit Rewards
• Agent-RLVR: Training Software Engineering Agents via Guidance and Environment Rewards
• Diffusion World Model: Future Modeling Beyond Step-by-Step Rollout for Offline Reinforcement Learning
• A Survey on Code Generation with LLM-based Agents
• Efficient-DLM: From Autoregressive to Diffusion Language Models, and Beyond in Speed
• Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding
• NExtLong: Toward Effective Long-Context Training without Long Documents
• Scaling Synthetic Data Creation with 1,000,000,000 Personas

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