Unified Trajectory Matching Policy Optimization: Diverse T2I Generation and VLA Generalization
cs.CV
Submitted: 2026-09-28
Updated: 2026-09-28
Code: https://github.com/black-forest-labs/flux
Terminology
Sources
- $\pi_\texttt{RL}$: Online RL Fine-tuning for Flow-based Vision-Language-Action Models
- GARDO: Reinforcing Diffusion Models without Reward Hacking
- Dynamic-TreeRPO: Breaking the Independent Trajectory Bottleneck with Structured Sampling
- OP-GRPO: Efficient Off-Policy GRPO for Flow-Matching Models
- RewardDance: Reward Scaling in Visual Generation
- Understanding Reward Hacking in Text-to-Image Reinforcement Learning
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- GRAPE: Generalizing Robot Policy via Preference Alignment
- Behavioral Mode Discovery for Fine-tuning Multimodal Generative Policies
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment
- Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis
- RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models