技术雷达TECH RADAR · GITHUB · ARXIV · HACKER NEWS

3D Gaussian avatars support fast rendering, however, their real-time animation is often challenged by the costly neural inference. We address this bottleneck and show that the animation of pretrained

Ramazan Fazylov, Stamatis Lefkimmiatis, Ivan Laptev

LLMs are increasingly applied to cybersecurity workflows, where they are expected to translate analysts' intent into tool invocations. However, existing evaluations focus on knowledge-based assessment

Pengfei Li, Naufal Suryanto, Sicheng Zhang

Building reliable robot capabilities across diverse tasks requires substantial human effort to develop and maintain skills, design rewards, and integrate perception with control. We present Reconstruc

Yen-Jen Wang, Haozhe Jiang, Shuying Deng

In diffusion transformers, a class label or a text prompt is embedded once, and the same condition is reused at every denoising step. We ask whether predicted embeddings can serve as this condition in

Sihan Xu, Ji Xie, Zilin Wang

What makes great scientists great? Even as AI systems start to make progress on open problems, scientists remain far ahead of them at sensing which prior idea, buried in an ever-growing archive of res

Sohyeon Kim, Yoonho Lee, Bo Liu

High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active structure and then synthesize local geometry. While effective, this design fragme

Tianjiao Yu, Xinzhuo Li, Yifan Shen

We show that multimodal models possess strong reasoning abilities and that an appropriate harness can unlock their potential to solve tasks across diverse interactive environments. We introduce VISTA,

Qiushi Han, Keya Hu, Linlu Qiu

Full-parameter fine-tuning of large language models (LLMs) incurs substantial optimizer state memory overhead, limiting the model sizes that fit on modern GPUs. Existing approaches either compress opt

Jichao Jiang, Cristian McGee, El Houcine Bergou

Several state-of-the-art methods for online reinforcement learning in continuous control improve policies using action gradients of a learned critic. However, critics are typically trained to predict

Sebastian Sanokowski, Alireza Sarmadi, Majid Khadiv

The generalized Schrödinger bridge on a graph moves mass between two distributions while charging a cost for the states visited. It has been approached by learning the rates of a controlled continuous

Akshay Balsubramani

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论

原文 · 讨论