Papers

Trending research and the full catalog - each paper linked to the benchmarks, methods, and models it introduces.

Filtered by domain: image-generationClear

Are We Overthinking One-Step Generation?

14 Oct 2026

Diffusion and flow matching are now dominant frameworks for generative modeling in continuous domains, but they typically require multiple network evaluations at inference.

Image generation Text-to-Image Generation

640.2/h

MEND: RL For Flow Models via Proximal Velocity Matching

5 Oct 2026

Reward post-training of flow models either reweights the model's own samples under a KL penalty or a frozen reference, often for thousands of updates, or backpropagates the reward and moves every sample without checking that the move is worth its size.

Image generation Reinforcement Learning

50.0/h

Filtering here covers the 2,000 most recent papers, as much as one page can hold in memory. See the full index of 22,656 papers.