Hanrong Ye

BaT: Towards Self-Evolving Medical Research Agent with Stage Rubrics

Junqi Liu, Yufan He, Yexiao He, Pengfei Guo et autres

Long-horizon agents are beginning to automate complete workflows that produce code, reports, and research artifacts. Medical imaging workflows are multi-stage and data-sensitive, while expert trajectories remain scarce and difficult to share. Structured benchmarks can localize failures through stage-level rubrics, but standard post-training …

Audio-Visual Flamingo: Open Audio-Visual Intelligence for Long and Complex Videos

Sreyan Ghosh, Arushi Goel, K. S. Jayakumar, Lasha Koroshinadze et autres

We present Audio-Visual Flamingo (AV-Flamingo), a fully open state-of-the-art audio-visual large language model (AV-LLM) for joint understanding and reasoning over audio, images, and long-form videos. Unlike prior AV-LLMs that primarily focus on short clips, AV-Flamingo is designed for understanding and reasoning over …

Audio-Visual Flamingo: Open Audio-Visual Intelligence for Long and Complex Videos

Sreyan Ghosh, Arushi Goel, K. S. Jayakumar, Lasha Koroshinadze et autres

We present Audio-Visual Flamingo (AV-Flamingo), a fully open state-of-the-art audio-visual large language model (AV-LLM) for joint understanding and reasoning over audio, images, and long-form videos. Unlike prior AV-LLMs that primarily focus on short clips, AV-Flamingo is designed for understanding and reasoning over …

gb, us (code pays fourni par la source)

Nemotron-Labs-Diffusion-Image: Advancing Masked Discrete Diffusion for High-Resolution Image Synthesis

Shufan Li, Greg Heinrich, Hanrong Ye, Yonggan Fu et autres

We propose Nemotron-Labs-Diffusion-Image, a state-of-the-art masked discrete diffusion model (MDM) for high-resolution text-to-image synthesis. Compared with prior work on masked image generation, Nemotron-Labs-Diffusion-Image addresses two key challenges. First, unlike continuous diffusion models which progressively refine latent representations across the entire image, standard …

Nemotron-Labs-Diffusion-Image: Advancing Masked Discrete Diffusion for High-Resolution Image Synthesis

Shufan Li, Greg Heinrich, Hanrong Ye, Yonggan Fu et autres

We propose Nemotron-Labs-Diffusion-Image, a state-of-the-art masked discrete diffusion model (MDM) for high-resolution text-to-image synthesis. Compared with prior work on masked image generation, Nemotron-Labs-Diffusion-Image addresses two key challenges. First, unlike continuous diffusion models which progressively refine latent representations across the entire image, standard …

Vesta: A Generalist Embodied Reasoning Model

Johan Björck, Zhiqi Li, Yunze Man, Jing Wang et autres

Robots operating in open-world environments must seamlessly integrate localization, spatial reasoning, navigation, and long-horizon planning. While specialist models excel at individual tasks, deploying a multi-model stack is computationally expensive and prone to cascading errors. We present Vesta, a unified embodied generalist that …

Vesta: A Generalist Embodied Reasoning Model

Johan Björck, Zhiqi Li, Yunze Man, Yi-Xiang Wang et autres

Robots operating in open-world environments must seamlessly integrate localization, spatial reasoning, navigation, and long-horizon planning. While specialist models excel at individual tasks, deploying a multi-model stack is computationally expensive and prone to cascading errors. We present Vesta, a unified embodied generalist that …

us, hk, sg, de (code pays fourni par la source)

Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders

Yitong Jiang, Hongjun Wang, Collin McCarthy, Hanrong Ye et autres

Vision foundation models are bottlenecked by the quadratic cost of self-attention, which limits usable resolution and increases the cost of large-scale pretraining. Subquadratic alternatives such as linear attention and state-space models reduce this cost, but often serialize images into 1D token streams …

Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders

Yitong Jiang, Hongjun Wang, Collin McCarthy, Hanrong Ye et autres

Vision foundation models are bottlenecked by the quadratic cost of self-attention, which limits usable resolution and increases the cost of large-scale pretraining. Subquadratic alternatives such as linear attention and state-space models reduce this cost, but often serialize images into 1D token streams …

hk, us (code pays fourni par la source)

JetViT: Efficient High-Resolution Vision Transformer with Post-Training Attention Search

Dongyun Zou, Zhuoyang Zhang, Junyu Chen, Wenkun He et autres

We introduce JetViT, a novel family of hybrid-architecture Vision Transformer (ViT) models that match the accuracy of state-of-the-art full-attention vision foundation models while achieving substantially higher inference efficiency on high-resolution images. At the core of our approach is Post-Training Attention Search, a …

JetViT: Efficient High-Resolution Vision Transformer with Post-Training Attention Search

Dongyun Zou, Zhuoyang Zhang, Junyu Chen, Wenkun He et autres

We introduce JetViT, a novel family of hybrid-architecture Vision Transformer (ViT) models that match the accuracy of state-of-the-art full-attention vision foundation models while achieving substantially higher inference efficiency on high-resolution images. At the core of our approach is Post-Training Attention Search, a …

Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence

NVIDIA, :, Amala Sanjay Deshmukh, Kateryna Chumachenko et autres

We introduce Nemotron 3 Nano Omni, the latest model in the Nemotron multimodal series and the first to natively support audio inputs alongside text, images, and video. Nemotron 3 Nano Omni delivers consistent accuracy improvements over its predecessor, Nemotron Nano V2 VL, …