I am a PhD student at the University of California, San Diego, advised by Prof. Xiaolong Wang. During my PhD studies, I interned at NVIDIA and Adobe, and my research has been supported by Qualcomm Innovation Fellowship. Prior to my PhD, I earned my Master’s and Bachelor’s degrees in computer science from National Tsing Hua University.
Research directions
I'm interested in building multimodal foundation models that understand space, act intelligently, and self-evolve through real-world experience.
News
Gave talks on “Grounding 3D Reasoning and Long-Horizon Actions in VLMs and VLAs” at the CVPR 2026 Workshop on Efficient Deep Learning for Computer Vision () and Foundation Models Meet Embodied Agents in Denver. Thank you for having me!
Excited to share Cosmos 3, NVIDIA's omnimodal world foundation model powering the next generation of Physical AI — grateful to have contributed to its spatial reasoning capabilities.
GR3D is accepted by CVPR 2026.
Earlier updates
Released code for SR-3D, together with SR-3D-Bench.
We’ve open-sourced the NaVILA framework. Including the VLA, locomotion policy, and the navigation benchmark.
SpatialRGPT was demoed at GTC 2025 as
a
part of Agentic AI for Physical
Operations!
Publications & preprints
Full list ↗Long-Horizon Manipulation via Trace-Conditioned VLA Planning
CoRL, 2026
Long-horizon manipulation via a task-management VLM with visual trace conditioning.
Long-horizon manipulation remains challenging for vision-language-action (VLA) policies: real tasks are multi-step, progress-dependent, and brittle to compounding execution errors. We present LoHo-Manip, a modular framework that scales short-horizon VLA execution to long-horizon instruction following via a dedicated task-management VLM. The manager is decoupled from the executor and is invoked in a receding-horizon manner: given the current observation, it predicts a progress-aware remaining plan that combines (i) a subtask sequence with an explicit done + remaining split as lightweight language memory, and (ii) a visual trace -- a compact 2D keypoint trajectory prompt specifying where to go and what to approach next. The executor VLA is adapted to condition on the rendered trace, thereby turning long-horizon decision-making into repeated local control by following the trace. Crucially, predicting the remaining plan at each step yields an implicit closed loop: failed steps persist in subsequent outputs, and traces update accordingly, enabling automatic continuation and replanning without hand-crafted recovery logic or brittle visual-history buffers. Extensive experiments spanning embodied planning, long-horizon reasoning, trajectory prediction, and end-to-end manipulation in simulation and on a real Franka robot demonstrate strong gains in long-horizon success, robustness, and out-of-distribution generalization.
Grounded 3D-Aware Spatial Vision-Language Modeling
CVPR, 2026
Unified Spatial Reasoning & 3D Grounding VLMs with visual CoT (Thinking with Regions).
GR3D is a spatial vision language model equipped with three complementary grounding capabilities — explicit 2D grounding, implicit 2D grounding, and monocular 3D grounding — within a single framework. GR3D introduces an implicit grounding mechanism that identifies entity mentions during generation and inserts the corresponding region tokens into the text stream, allowing the model to reference visual evidence on the fly when producing spatial chain-of-thought responses. Together, these grounding capabilities enable GR3D to decompose complex spatial understanding problems into grounded 2D perception followed by 3D inference.
OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM
ICLR, 2026
NVIDIA's state-of-the-art 9B Omni-Modal LLMs.
Advancing machine intelligence requires developing the ability to perceive across multiple modalities, much as humans sense the world. We introduce OmniVinci, an initiative to build a strong, open-source, omni-modal LLM. We carefully study the design choices across model architecture and data curation. For model architecture, we present three key innovations: (i) OmniAlignNet for strengthening alignment between vision and audio embeddings in a shared omni-modal latent space; (ii) Temporal Embedding Grouping for capturing relative temporal alignment between vision and audio signals; and (iii) Constrained Rotary Time Embedding for encoding absolute temporal information in omni-modal embeddings. We introduce a curation and synthesis pipeline that generates 24M single-modal and omni-modal conversations. We find that modalities reinforce one another in both perception and reasoning. Our model, OmniVinci, outperforms Qwen2.5-Omni with +19.05 on DailyOmni (cross-modal understanding), +1.7 on MMAR (audio), and +3.9 on Video-MME (vision), while using just 0.2T training tokens - a 6 times reduction compared to Qwen2.5-Omni's 1.2T. We finally demonstrate omni-modal advantages in downstream applications spanning robotics, medical AI, and smart factory.
3D Aware Region Prompted Vision Language Model
ICLR, 2026
Region-level spatial reasoning for both single-view and multi-view inputs.
We present Spatial Region 3D (SR-3D) aware vision-language model that connects single-view 2D images and multi-view 3D
data through a shared visual token space. SR-3D supports flexible region prompting, allowing users to annotate regions
with bounding boxes, segmentation masks on any frame, or directly in 3D, without the need for exhaustive multi-frame
labeling. We achieve this by enriching 2D visual features with 3D positional embeddings, which allows the 3D model to
draw upon strong 2D priors for more accurate spatial reasoning across frames, even when objects of interest do not
co-occur within the same view. Extensive experiments on both general 2D vision language and specialized 3D spatial
benchmarks demonstrate that SR-3D achieves state-of-the-art performance, underscoring its effectiveness for unifying 2D
and 3D representation space on scene understanding. Moreover, we observe applicability to in-the-wild videos without
sensory 3D inputs or ground-truth 3D annotations, where SR-3D accurately infers spatial relationships and metric
measurements.
EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos
IROS, 2026
Robust dexterous manipulation generalist model utilizing diverse egocentric human manipulation videos.
Real robot data collection for imitation learning has led to significant advancements in robotic manipulation. However, the requirement for robot hardware in the process fundamentally constrains the scale of the data. In this paper, we explore training Vision-Language-Action (VLA) models using egocentric human videos. The benefit of using human videos is not only for their scale but more importantly for the richness of scenes and tasks. With a VLA trained on human video that predicts human wrist and hand actions, we can perform Inverse Kinematics and retargeting to convert the human actions to robot actions. We fine-tune the model using a few robot manipulation demonstrations to obtain the robot policy, namely EgoVLA. We propose a simulation benchmark called Ego Humanoid Manipulation Benchmark, where we design diverse bimanual manipulation tasks with demonstrations. We fine-tune and evaluate EgoVLA with Ego Humanoid Manipulation Benchmark and show significant improvements over baselines and ablate the importance of human data.
NaVILA: Legged Robot Vision-Language-Action Model for Navigation
RSS, 2025
A two-level framework that combines VLAs with locomotion skills for navigation. The VLA is adapted from a VLM and learns from human touring videos.
NVILA: Efficient Frontier Visual Language Models
CVPR, 2025
Efficient frontier VLM models with efficient training and inference.
Visual language models (VLMs) have made significant
advances in accuracy in recent years. However, their efficiency has
received much less attention. This paper introduces NVILA, a family of open VLMs designed to optimize both
efficiency
and accuracy. Building on top of VILA, we improve its model architecture by first scaling up the spatial and
temporal
resolutions, and then compressing visual tokens. This "scale-then-compress" approach enables NVILA to
efficiently
process high-resolution images and long videos. We also conduct a systematic investigation to enhance the
efficiency of
NVILA throughout its entire lifecycle, from training and fine-tuning to deployment. NVILA matches or surpasses
the
accuracy of many leading open and proprietary VLMs across a wide range of image and video benchmarks. At the
same time,
it reduces training costs by 4.5X, fine-tuning memory usage by 3.4X, pre-filling latency by 1.6-2.2X, and
decoding
latency by 1.2-2.8X. We will soon make our code and models available to facilitate reproducibility.
SpatialRGPT: Grounded Spatial Reasoning in Vision-Language Models
NeurIPS, 2024
A powerful region-level VLM adept at 3D spatial reasoning.
✨ Demoed at GTC 2025 as a part of Agentic AI for Physical Operations!
Vision Language Models (VLMs) have
demonstrated remarkable performance in 2D vision and language tasks. However, their ability to
reason about spatial arrangements remains limited. In this work, we introduce Spatial Region GPT
(SpatialRGPT) to enhance VLMs' spatial perception and reasoning capabilities. SpatialRGPT
advances VLMs' spatial understanding through two key innovations: (1) a data curation pipeline
that enables effective learning of regional representation from 3D scene graphs, and (2) a
flexible plugin module for integrating depth information into the visual encoder of existing
VLMs. During inference, when provided with user-specified region proposals, SpatialRGPT can
accurately perceive their relative directions and distances. Additionally, we propose
SpatialRGBT-Bench, a benchmark with ground-truth 3D annotations encompassing indoor, outdoor,
and simulated environments, for evaluating 3D spatial cognition in VLMs. Our results demonstrate
that SpatialRGPT significantly enhances performance in spatial reasoning tasks, both with and
without local region prompts. The model also exhibits strong generalization capabilities,
effectively reasoning about complex spatial relations and functioning as a region-aware dense
reward annotator for robotic tasks.
TUVF: Learning Generalizable Texture UV Radiance Fields
ICLR, 2024
Learning generalizable texture UV radiance fields for shapes.
Textures are a vital aspect of creating
visually
appealing and realistic 3D models. In this paper, we study the problem of generating
high-fidelity
texture given shapes of 3D assets, which has been relatively less explored compared with generic
3D
shape modeling. Our goal is to facilitate a controllable texture generation process, such that
one
texture code can correspond to a particular appearance style independent of any input shapes
from a
category. We introduce Texture UV Radiance Fields (TUVF) that generate textures in a learnable
UV sphere
space rather than directly on the 3D shape. This allows the texture to be disentangled from the
underlying shape and transferable to other shapes that share the same UV space, i.e., from the
same
category. We integrate the UV sphere space with the radiance field, which provides a more
efficient and
accurate representation of textures than traditional texture maps. We perform our experiments on
real-world object datasets where we achieve not only realistic synthesis, but also substantial
improvements over state-of-the-arts on texture controlling and editing.
Autoregressive 3D Shape Generation via Canonical Mapping
ECCV, 2022
We decompose the point cloud into meaningful shape sequences, then we encode these sequences through a transformer for generation.
With the capacity of modeling long-range
dependencies in sequential data, transformers have shown remarkable performances in a variety of
generative tasks such as image, audio, and text generation. Yet, taming them in generating less
structured and voluminous data formats such as high-resolution point clouds have seldom been
explored
due to ambiguous sequentialization processes and infeasible computation burden. In this paper,
we aim to
further exploit the power of transformers and employ them for the task of 3D point cloud
generation. The
key idea is to decompose point clouds of one category into semantically aligned sequences of
shape
compositions, via a learned canonical space. These shape compositions can then be quantized and
used to
learn a context-rich composition codebook for point cloud generation. Experimental results on
point
cloud reconstruction and unconditional generation show that our model performs favorably against
state-of-the-art approaches. Furthermore, our model can be easily extended to multi-modal shape
completion as an application for conditional shape generation.
Learning 3D Dense Correspondence via Canonical Point Autoencoder
NeurIPS, 2021
Unsupervised learning of dense 3D correspondence.
We propose a canonical point
autoencoder
(CPAE) that predicts dense correspondences between 3D shapes of the same category. The
autoencoder
performs two key functions: (a) encoding an arbitrarily ordered point cloud to a canonical
primitive,
e.g., a sphere, and (b) decoding the primitive back to the original input instance shape. As
being
placed in the bottleneck, this primitive plays a key role to map all the unordered point clouds
on the
canonical surface and to be reconstructed in an ordered fashion. Once trained, points from
different
shape instances that are mapped to the same locations on the primitive surface are determined to
be a
pair of correspondence. Our method does not require any form of annotation or self-supervised
part
segmentation network and can handle unaligned input point clouds. Experimental results on 3D
semantic
keypoint transfer and part segmentation transfer show that our model performs favorably against
state-of-the-art correspondence learning methods.
Technical reports
Vesta: A Generalist Embodied Reasoning Model
NVIDIA Technical Report, 2026
A state-of-the-art Robot System 2 embodied-reasoning VLM that unifies localization, navigation, memory, reasoning, tool use, and long-horizon planning.
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 consolidates these capabilities into a single foundation model. Our approach combines a diverse and massive curated corpus designed to induce spatial grounding and a simple multimodal memory harness that enables reasoning over extended time horizons. Across diverse benchmarks, Vesta on average beats individual SOTA baselines by >20% and beats an ensemble of per-category-best baselines by >10% — thus demonstrating that a generalist model can match or exceed specialists. On real-world robotic tasks requiring memory and reasoning, Vesta improves task success by >35%. Our work thus demonstrates that a single generalist is a feasible, scalable, and arguably preferable alternative to combining specialists.
Cosmos 3: Omnimodal World Models for Physical AI
NVIDIA Technical Report, 2026
NVIDIA's omnimodal world foundation model that unifies understanding, generation, simulation, and action across text, image, video, audio, and robot actions for Physical AI.
We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-transformers architecture. By supporting highly flexible input-output configurations, Cosmos 3 seamlessly unifies critical modalities for Physical AI—effectively subsuming vision-language models, video generators, world simulators, and world-action models into a single framework. Our evaluation demonstrates that Cosmos 3 establishes a new state-of-the-art across a diverse suite of understanding and generation tasks, demonstrating omnimodal world models as scalable, general-purpose backbones for embodied agents. Our post-trained Cosmos 3 models were ranked as the best open-source Text-to-Image and Image-to-Video models by Artificial Analysis, and the best policy model by RoboArena at the time the technical report was written.