Internalizing Agent Experience into Diffusion Model Weights via On-Policy Context Distillation

arXiv:2610.07250 · cs.AI · Submitted 2026-10-05 · Read on arXiv

cs.AI

Submitted: 2026-10-05

Updated: 2026-10-05

Terminology

• Diffusion Model — 1× in this paper · explained in 9 episodes
• Distillation — 1× in this paper · explained in 2 episodes

Sources

• Flow-OPD: On-Policy Distillation for Flow Matching Models
• GenAgent: Scaling Text-to-Image Generation via Agentic Multimodal Reasoning
• GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation
• Continual Harness: Online Adaptation for Self-Improving Foundation Agents
• CRAFT: Continuous Reasoning and Agentic Feedback Tuning for Multimodal Text-to-Image Generation
• Flow-GRPO: Training Flow Matching Models via Online RL
• WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation
• Continual Diffusion: Continual Customization of Text-to-Image Diffusion with C-LoRA
• ImAgent: A Unified Multimodal Agent Framework for Test-Time Scalable Image Generation
• SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning
• MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild
• On-Policy Context Distillation for Language Models
• MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
• Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation

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