57 citations · 157 across the 32 of their papers we have counts for
34 papers
Getting Space Right: Agentic Rewards for Image Generation
Xiaofeng Tan, Jiaxin Lin, Haobo Yuan +5
Reinforcement learning (RL) is promising for spatial text-to-image generation, but existing work largely focuses on policy optimization, assuming accurate rewards and effective onl…
Task-State Adaptation with Prototype Memory for Multi-Task Dense Prediction
Yangyang Xu, Haobo Yuan, Yuzhu Wang +4
Vision foundation backbones provide strong representations for dense prediction, yet a single shared feature still needs to support tasks with different, image-dependent adaptation…
Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment
Han-Jun Ko, Jr-Jen Chen, Haobo Yuan +4
Vision-language models (VLMs) struggle to generalize in interactive physical reasoning, particularly under unseen tasks and environments. Two key failure modes are prominent: hallu…
PixelEyes: Decoupling Perception and Reasoning for Pinpoint Visual Evidence Seeking
Dengxian Gong, Yuanzheng Wu, Haobo Yuan +11
This paper explores multi-turn visual reasoning and observes that MLLMs repeatedly fail to localize the target, leading to long, redundant trajectories. We attribute this failure t…
Cosmos 3: Omnimodal World Models for Physical AI
NVIDIA, :, Aditi +293
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-t…
Report of the 5th PVUW Challenge: Towards More Diverse Modalities in Pixel-Level Understanding
Chang Liu, Henghui Ding, Nikhila Ravi +40
This report summarizes the objectives, datasets, and top-performing methodologies of the 2026 Pixel-level Video Understanding in the Wild (PVUW) Challenge, hosted at CVPR 2026, whi…
SaSaSaSa2VA: 2nd Place of the 5th PVUW MeViS-Text Track
Dengxian Gong, Quanzhu Niu, Shihao Chen +6
Referring video object segmentation (RVOS) commonly grounds targets in videos based on static textual cues. MeViS benchmark extends this by incorporating motion-centric expressions…
SAMTok: Representing Any Mask with Two Words
Yikang Zhou, Tao Zhang, Dengxian Gong +13
Pixel-wise capabilities are essential for building interactive intelligent systems. However, pixel-wise multi-modal LLMs (MLLMs) remain difficult to scale due to complex region-lev…
Visual Reasoning Tracer: Object-Level Grounded Reasoning Benchmark
Haobo Yuan, Yueyi Sun, Yanwei Li +7
Recent advances in Multimodal Large Language Models (MLLMs) have significantly improved performance on tasks such as visual grounding and visual question answering. However, the re…
The 1st Solution for 7th LSVOS RVOS Track: SaSaSa2VA
Quanzhu Niu, Dengxian Gong, Shihao Chen +6
Referring video object segmentation (RVOS) requires segmenting and tracking objects in videos conditioned on natural-language expressions, demanding fine-grained understanding of b…
LSVOS 2025 Challenge Report: Recent Advances in Complex Video Object Segmentation
Chang Liu, Henghui Ding, Kaining Ying +46
This report presents an overview of the 7th Large-scale Video Object Segmentation (LSVOS) Challenge held in conjunction with ICCV 2025. Besides the two traditional tracks of LSVOS…
ParaCook: On Time-Efficient Planning for Multi-Agent Systems
Shiqi Zhang, Xinbei Ma, Yunqing Xu +7
Large Language Models (LLMs) exhibit strong reasoning abilities for planning long-horizon, real-world tasks, yet existing agent benchmarks focus on task completion while neglecting…
DenseWorld-1M: Towards Detailed Dense Grounded Caption in the Real World
Xiangtai Li, Tao Zhang, Yanwei Li +13
Multimodal Large Language Models (MLLMs) demonstrate a complex understanding of scenes, benefiting from large-scale and high-quality datasets. Most existing caption datasets lack t…
PVUW 2025 Challenge Report: Advances in Pixel-level Understanding of Complex Videos in the Wild
Henghui Ding, Chang Liu, Nikhila Ravi +33
This report provides a comprehensive overview of the 4th Pixel-level Video Understanding in the Wild (PVUW) Challenge, held in conjunction with CVPR 2025. It summarizes the challen…
An Empirical Study of GPT-4o Image Generation Capabilities
Sixiang Chen, Jinbin Bai, Zhuoran Zhao +16
The landscape of image generation has rapidly evolved, from early GAN-based approaches to diffusion models and, most recently, to unified generative architectures that seek to brid…
4th PVUW MeViS 3rd Place Report: Sa2VA
Haobo Yuan, Tao Zhang, Xiangtai Li +5
Referring video object segmentation (RVOS) is a challenging task that requires the model to segment the object in a video given the language description. MeViS is a recently propos…
Sa2VA: Marrying SAM2 with MLLM for Dense Grounded Understanding of Images and Videos
Haobo Yuan, Xiangtai Li, Tao Zhang +8
This work presents Sa2VA, the first comprehensive, unified model for dense grounded understanding of both images and videos. Unlike existing multi-modal large language models, whic…
LLAVADI: What Matters For Multimodal Large Language Models Distillation
Shilin Xu, Xiangtai Li, Haobo Yuan +3
The recent surge in Multimodal Large Language Models (MLLMs) has showcased their remarkable potential for achieving generalized intelligence by integrating visual understanding int…
Mamba or RWKV: Exploring High-Quality and High-Efficiency Segment Anything Model
Haobo Yuan, Xiangtai Li, Lu Qi +4
Transformer-based segmentation methods face the challenge of efficient inference when dealing with high-resolution images. Recently, several linear attention architectures, such as…
OMG-LLaVA: Bridging Image-level, Object-level, Pixel-level Reasoning and Understanding
Tao Zhang, Xiangtai Li, Hao Fei +5
Current universal segmentation methods demonstrate strong capabilities in pixel-level image and video understanding. However, they lack reasoning abilities and cannot be controlled…
Point Cloud Mamba: Point Cloud Learning via State Space Model
Tao Zhang, Haobo Yuan, Lu Qi +5
Recently, state space models have exhibited strong global modeling capabilities and linear computational complexity in contrast to transformers. This research focuses on applying s…
OMG-Seg: Is One Model Good Enough For All Segmentation?
Xiangtai Li, Haobo Yuan, Wei Li +6
In this work, we address various segmentation tasks, each traditionally tackled by distinct or partially unified models. We propose OMG-Seg, One Model that is Good enough to effici…
RMP-SAM: Towards Real-Time Multi-Purpose Segment Anything
Shilin Xu, Haobo Yuan, Qingyu Shi +9
Recent segmentation methods, which adopt large-scale data training and transformer architecture, aim to create one foundation model that can perform multiple tasks. However, most o…
Open-Vocabulary SAM: Segment and Recognize Twenty-thousand Classes Interactively
Haobo Yuan, Xiangtai Li, Chong Zhou +3
The CLIP and Segment Anything Model (SAM) are remarkable vision foundation models (VFMs). SAM excels in segmentation tasks across diverse domains, whereas CLIP is renowned for its…