Generative media research

The frontier

Joint deterministic-generative model extends reliable extreme-precipitation nowcasting to six hours

The authors present MW-Nowcast, a six-hour ensemble radar nowcasting model that jointly learns a deterministic predictor for organized precipitation structure and a generative flow-matching model for local residuals. The model doubles the available warning time for the most intense rainfall, delivering 6 h forecasts with skill previously confined to 3 h for the leading generative baseline.

Ning Wang, Zuliang Fang, Weixin Jin +19

Generative media

Adaptive reference attention reuse speeds up conditioned diffusion transformers

The authors propose RefAdapt-DiT, a training-free framework that adaptively controls joint attention computation between reference images and target generation in diffusion transformers. By monitoring reference drift and target-to-reference attention, it selectively reuses stale reference states, achieving speedups of up to 3.54× on image editing with minimal quality loss.

today

Generative media

Three-stage latent generator reaches 1.11 FID on ImageNet

Xu and Wang train a few-step image generator by first imitating a pretrained teacher, then optimizing the entire sequence of latent transitions against real ImageNet data. Their three-stage model reports FID 1.11 and Inception Score 282, while training entirely in latent space and without directly minimizing Fréchet distance.

today

Speech & audio

YuE2 unifies symbolic and audio music generation via explicit score planning

YuE2 unifies symbolic and audio music generation by first writing an explicit score, then expanding it into semantic tokens and synthesizing full-song audio. Expert evaluations show that symbolic planning improves quality, and YuE2 achieves state-of-the-art results on benchmarks.

today

Generative media

Fine-grained routing makes sparse video attention faster without quality loss

Zhang et al. introduce VSA2, a trainable sparse-attention system for video diffusion transformers that selects a small, query-dependent subset of token interactions. The authors report substantially faster 720p generation than a FlashAttention-3 baseline, while maintaining comparable or better video quality.

today

Generative media

High-fidelity 3D texture models can train on images alone

Wang and colleagues present Tex-Zero, a native 3D texture system trained exclusively on 2D images rather than textured 3D assets. They convert image patches into artificial 3D arrangements, then use them to train both a variational autoencoder and a diffusion transformer. The authors report that the resulting model reconstructs and generates detailed textures on real, previously unseen 3D geometry.

today

Generative media

Projecting critic errors stabilizes four-step video diffusion distillation

Wang et al. identify inaccurate critic estimates as a source of oversaturation and artifacts during Distribution Matching Distillation, a method for compressing diffusion generation into a few denoising evaluations. Their proposed projection filters the error-prone component of each update, improving stability and benchmark performance in four-step video and joint video-audio generation.

today

All Generative media papers →

Every paper read and written up by the research desk from the daily arXiv scan · threads are maintained lines of inquiry with running syntheses