Arxiv

In this work, we present Next-Frame Diffusion (NFD), an autoregressive diffusion transformer that incorporates block-wise causal attention, enabling iterative sampling and efficient inference via parallel token generation within each frame.

We present ChemAgent, a novel framework designed to improve the performance of LLMs through a dynamic, self-updating library.