RL Post-Training in Diffusion Transformers Is Activation Steering
Abstract
A fixed set of steering vectors added to a frozen diffusion transformer recovers most of the quality gains from reinforcement learning post-training, indicating that much of the improvement comes from a constant shift rather than prompt-specific transformations.
Reinforcement learning (RL) post-training substantially improves the visual quality, text rendering, and compositional accuracy of diffusion transformers. Because capabilities such as rendering requested text or composing multiple objects are inherently prompt-dependent, a natural hypothesis is that improving them requires learning complex, input-dependent transformations. To isolate how much of these gains an input-independent shift explains, we probe each RL-tuned model with a deliberately simple parameterization: a fixed set of steering vectors added to the block outputs of the frozen base model. On SD3.5-M, these vectors hold 65x fewer parameters than the RL-tuned LoRA (0.29M against 18.8M). We fit them to match the velocity predictions of public RL-tuned checkpoints, with no reward in the objective. Although the vectors are identical for every prompt, token, and sampling seed, they recover 90.0-96.3% of the reward gains of six RL-tuned image models, and 89.8% and 111.9% on two RL-tuned video models. Initialized at zero and trained with online RL, the vectors alone exceed the reward gains of weight-based RL on text rendering (OCR) and human preference (PickScore) on SD3.5-M. A constant shift thus suffices for most of the gain of policy-gradient RL post-training: without new input-dependent transformations, it steers the pretrained generator toward what it can already compute.
Community
RL post-training of diffusion transformers is mostly a fixed shift. Fixed steering vectors on the block outputs of the frozen base model, identical for every prompt, token and seed, recover 90.0–96.3% of the reward gain of six RL-tuned image models, with 65× fewer parameters than the RL-tuned LoRA on SD3.5-M.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- DualHeads: Reward-tilted few-step diffusion made simple (2026)
- Routing Policy Optimization for Online Diffusion Reinforcement Learning (2026)
- The Surprising Effectiveness of Deleting Weights in LLM Post-Training (2026)
- ExploreNet: Learning Where to Explore in Diffusion GRPO (2026)
- Steering Vectors as a Training Signal in LLM Post-Training (2026)
- LongLive-Plug: Once-for-All Distillation for Video Generation (2026)
- Behavioral Steering of Vision-Language-Action Models in the Low-Parameter Regime (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2611.26627 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper