<div class="news">
<h2>news</h2>
<div class="table-responsive" >
<table class="table table-sm table-borderless">
<tr>
<th scope="row">Sep 24, 2025</th>
<td>
We’re excited to announce that our new preprint <a href="https://arxiv.org/abs/2510.21986">Sprint: Sparse-Dense Residual Fusion for Efficient Diffusion Transformers</a> is now available on arXiv. Our proposed <b>SPRINT</b> 🚀 framework achieves up to 9.8× faster DiT training while maintaining high generation quality.
</td>
</tr>
<tr>
<th scope="row">Sep 18, 2025</th>
<td>
Our <a href="https://arxiv.org/pdf/2510.21167">Blockwise Flow Matching</a> has been accepted to NeurIPS 2025! 🎉 <b>BFM</b> reduces the inference cost of standard DiTs by up to 4.9×, while maintaining comparable generation quality.
</td>
</tr>
<tr>
<th scope="row">May 31, 2025</th>
<td>
Excited to share that I’ve started my Research Internship at 👻 <b>Snap Inc</b>., joining the Creative Vision Team! I’ll be working on efficiency of generative AI and video diffusion models.
</td>
</tr>
</table>
</div>
</div>