Xiaosong Jia

LiFi: LiDAR Generation from Multi-View Images via Geometric and Semantic Collaborative Guidance

Sizhuo Zhou, Xiaosong Jia, Fanrui Zhang +10

Generating LiDAR point clouds from multi-view camera images is a valuable task with applications in controllable data synthesis, cross-modal simulation and various perception tasks…

CrossPortrait: Cross-Species Portrait Animation via Adversarial Motion Alignment

Gezhong Pan, Zhe Li, Yunfan Tao +7

Cross-species portrait animation, which aims to animate a source portrait of species A using a driving video of species B, is particularly challenging due to substantial anatomical…

Ego4WAM: What Matters When Scaling Egocentric Human Data for Robot Learning?

Zhihao Sun, Liu Liu, Xinjiang Wang +6

Egocentric human data provides a scalable source of experience for robot learning, but varies substantially in human-robot alignment, behavioral coverage, and available supervision…

HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing

Zhenjie Yang, Xingyu Jiao, Guopeng Zhong +18

Robotic manipulation with dexterous hands is a cornerstone of Embodied AI, yet its progress is stifled by the high cost of collecting embodiment-aware teleoperation data. While abu…

EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstration

Zuhao Ge, Yuchen Zhou, Weitao Zhou +8

Robust embodied robots should be able to recover from failures and retry tasks in order to operate reliably in unstructured and noisy real-world environments. Achieving this capabi…

Enhancing Video Physical Consistency via Role-aware Joint Training and Modality-decoupled Denoising

Guangting Zheng, Haojing Chen, Hao Li +6

While modern video diffusion models excel in visual fidelity, maintaining long-range physical consistency remains a formidable challenge. Conventional pixel-reconstruction objectiv…

Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation

Songbur Wong, Xiaosong Jia, Junqi You +12

Evaluating end-to-end autonomous driving (E2E-AD) remains challenging, as existing driving simulation methods often trade off closed-loop interactivity (e.g., CARLA) and real-world…

ReactSim-Bench: Benchmarking Reactive Behavior World Model Simulation in Autonomous Driving

Zhiyuan Zhang, Yanlun Peng, Jianing Zhang +7

Reactive capability is a key property of data-driven behavior world model simulators for autonomous driving simulation systems. With this capability, simulated world agents can res…

UniDexTok: A Unified Dexterous Hand Tokenizer from Real Data

Dong Fang, Youjun Wu, Yuanxin Zhong +4

Dexterous hands are essential for fine-grained manipulation, but their hardware designs vary substantially across embodiments. Differences in kinematics, joint definitions, and deg…

EvoMemNav: Efficient Self-Evolving Fine-Grained Memory for Zero-Shot Embodied Navigation

Zuhao Ge, Xiaosong Jia, Chao Wu +3

Building memory is essential for long-horizon planning in zero-shot embodied navigation. Detector-centric scene graphs often compress observations into sparse nodes, discarding fin…

GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization

Xiaosong Jia, Bowen Yang, Zuhao Ge +17

Vision-Language-Action (VLA) models aim for general robot learning by aligning action as a modality within powerful Vision-Language Models (VLMs). Existing VLAs rely on end-to-end…

Resolving Representation Ambiguity in Feedforward Novel View Synthesis Transformer via Semantic-Spatial Decoupling

Yihang Wu, Yihang Sun, Shaofeng Zhang +4

Transformer-based models have advanced feedforward novel view synthesis (NVS). Current architectures such as GS-LRM and LVSM mix semantic information (e.g., RGB) and spatial inform…

Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations

Zhiyuan Zhang, Zhenghao Jin, Yanlun Peng +8

Robustness is a critical requirement for deploying autonomous driving systems in the real world. Existing robustness benchmarks for autonomous driving have made important progress…

Attention Itself Could Retrieve.RetrieveVGGT: Training-Free Long Context Streaming 3D Reconstruction via Query-Key Similarity Retrieval

Zichen Zou, Xiaosong Jia, Zuxuan Wu +1

Visual Geometry Grounded Transformer (VGGT) advances 3D reconstruction via scalable Transformer architecture, but the quadratic complexity of global attention prevents long context…

SWIFT: Prompt-Adaptive Memory for Efficient Interactive Long Video Generation

Shanwen Tan, Hao Li, Jingtao Zhang +4

Streaming long-video generation faces a central challenge in continuous semantic switching, requiring adaptive memory to preserve coherent visual evolution. Current approaches rely…

Bench2Drive-VL: Benchmarks for Closed-Loop Autonomous Driving with Vision-Language Models

Xiaosong Jia, Yuqian Shao, Zhenjie Yang +3

With the rise of vision-language models (VLM), their application for autonomous driving (VLM4AD) has gained significant attention. Meanwhile, in autonomous driving, closed-loop eva…

Can Users Specify Driving Speed? Bench2Drive-Speed: Benchmark and Baselines for Desired-Speed Conditioned Autonomous Driving

Yuqian Shao, Xiaosong Jia, Langechuan Liu +1

End-to-end autonomous driving (E2E-AD) has achieved remarkable progress. However, one practical and useful function has been long overlooked: users may wish to customize the desire…

ACE-Brain-0: Spatial Intelligence as a Shared Scaffold for Universal Embodiments

Ziyang Gong, Zehang Luo, Anke Tang +21

Universal embodied intelligence demands robust generalization across heterogeneous embodiments, such as autonomous driving, robotics, and unmanned aerial vehicles (UAVs). However,…

PointAlign: Feature-Level Alignment Regularization for 3D Vision-Language Models

Yuanhao Su, Shaofeng Zhang, Xiaosong Jia +1

The development of 3D Vision-Language Models (VLMs), crucial for applications in robotics, autonomous driving, and augmented reality, is severely constrained by the scarcity of pai…

Efficient-LVSM: Faster, Cheaper, and Better Large View Synthesis Model via Decoupled Co-Refinement Attention

Xiaosong Jia, Yihang Sun, Junqi You +5

Feedforward models for novel view synthesis (NVS) have recently advanced by transformer-based methods like LVSM, using attention among all input and target views. In this work, we…

Repulsor: Accelerating Generative Modeling with a Contrastive Memory Bank

Shaofeng Zhang, Xuanqi Chen, Ning Liao +7

The dominance of denoising generative models (e.g., diffusion, flow-matching) in visual synthesis is tempered by their substantial training costs and inefficiencies in representati…

Spatial Retrieval Augmented Autonomous Driving

Xiaosong Jia, Chenhe Zhang, Yule Jiang +8

Existing autonomous driving systems rely on onboard sensors (cameras, LiDAR, IMU, etc) for environmental perception. However, this paradigm is limited by the drive-time perception…

DriveVGGT: Calibration-Constrained Visual Geometry Transformers for Multi-Camera Autonomous Driving

Xiaosong Jia, Yanhao Liu, Yu Hong +5

Feed-forward reconstruction has been progressed rapidly, with the Visual Geometry Grounded Transformer (VGGT) being a notable baseline. However, directly applying VGGT to autonomou…