Jing Tang, Tianyang Hu, Weijian Luo, Yihong Luo
We lifted 1 functions out of this paper's own repositories and ran 0 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| Luo-Yihong/TDM-R1 | — | 0 of 1 |
| Function | Status | Where it lives |
|---|---|---|
| sde_step_with_logprob | Not yet run | Luo-Yihong/TDM-R1/flow_grpo/diffusers_patch/sd3_sde_with_logprob.py pointer only (licence: NONE) · get_code("c3709c2daadf53d3") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
While few-step generative models have enabled powerful image and video generation at significantly lower cost, generic reinforcement learning (RL) paradigms for few-step models remain an unsolved problem. Existing RL approaches for few-step diffusion models strongly rely on back-propagating through differentiable reward models, thereby excluding the majority of important real-world reward signals, e.g., non-differentiable rewards such as humans' binary likeness, object counts, etc. To properly incorporate non-differentiable rewards to improve few-step generative models, we introduce TDM-R1, a novel reinforcement learning paradigm built upon a leading few-step model, Trajectory Distribution Matching (TDM). TDM-R1 decouples the learning process into surrogate reward learning and generator learning. Furthermore, we developed practical methods to obtain per-step reward signals along the deterministic generation trajectory of TDM, resulting in a unified RL post-training method that significantly improves few-step models' ability with generic rewards. We conduct extensive experiments ranging from textrendering, visual quality, and preference alignment. All results demonstrate that TDM-R1 is a powerful reinforcement learning paradigm for few-step text-to-image models, achieving state-of-the-art reinforcement learning performances on both in-domain and out-of-domain metrics. Furthermore, TDM-R1 also scales effectively to the recent strong Z-Image model, consistently outperforming both its 100-NFE and few-step variants with only 4 NFEs. Project page: https://github.com/Luo-Yihong/TDM-R1.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2603.07700")
get_code_for_paper("2603.07700")
have("2603.07700")
Connect an agent — have() is free.