- Rating
- 3
- Confidence
- 4
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
This paper studies when co-training the understanding and generation experts becomes redundant in unified multimodal RL. It shows that the understanding expert is mainly useful in the early training stage and proposes a staged strategy that freezes it once its contribution declines. The analysis is interesting and the method improves both efficiency and generation performance, but the evaluation is limited to MoT-style architectures and the freeze rule still depends on extra probes and calibrated thresholds. Overall, I find the main idea and observations interesting, and I would consider a higher score if the rebuttal can address these concerns well.
Critical strengths
-
The paper reveals that the benefit of jointly updating the understanding expert is strongly stage-dependent, and identifies a clear “front-loaded” contribution pattern during RL training.
-
The compensability and transport probes provide an interpretable way to characterize when co-training becomes redundant, and the staged strategy is validated across multiple backbones, reward models, and RL algorithms.
Critical weaknesses
-
Both evaluated backbones use MoT-style architectures with explicitly separated understanding and generation experts, so it is unclear whether the same redundancy pattern and freeze strategy hold for more tightly shared unified multimodal models.
-
Compensability and transport require additional velocity/Jacobian-based measurements, while the thresholds are calibrated on BAGEL before being transferred to SenseNova-U1. The practical overhead and robustness across substantially different models remain insufficiently explored.
Questions for authors
-
How well does the proposed freeze criterion generalize to unified multimodal models without explicitly separated understanding and generation experts?
-
What is the additional computational overhead of the compensability and transport probes during training, and is the resulting efficiency gain still significant after including this cost?
-
How sensitive is the method to the manually calibrated freeze thresholds, and would the same thresholds remain effective across larger model scales, different tasks, or substantially different RL settings?