Visual representations are largely developed for understanding, yet what makes them useful for generation remains unclear. We study this question in pixel diffusion transformers and uncover two key insights. First, better representation quality does not necessarily translate into greater generative utility, as external...
Zi-Hao Pan, Meng-Ting Chen, Cheng Shi et al.· 0 citations
With the rapid advancement of image editing models and their widespread application across various domains, there is an increasingly urgent need to deploy these model capabilities directly into real-world scenarios. However, existing benchmarks remain confined to simple single-image tasks, suffering from limited covera...
Qinye Zhou, Jun Zheng, Yongchao Du et al.· 2 citations
A tailored Bootstrapped Pareto Policy Optimization (BPPO) is proposed, which synergizes Bootstrapping Reward Rectification and Conflict-Aware Pareto Advantage Fusion (CPAF) and exposes critical reasoning deficits, highlighting imperative for knowledge-intensive benchmarks towards next-generation visual generation.
Yuan Wang, Yongchao Du, Mengting Chen et al.· arXiv.org· 1 citation
Swift-Image achieves leading aggregate performance among evaluated open-source models with only 6B parameters and 243K GPU training hours; the compressed 3B model incurs nearly no loss, while few-step distillation further improves aggregate editing performance with substantially fewer sampling steps.
