Peking University

university

https://www.pku.edu.cn/

Activity Feed Request to join this org

AI & ML interests

nlu

Recent Activity

xishushu submitted a paper 1 day ago

Know3D: Prompting 3D Generation with Knowledge from Vision-Language Models

zyh200727 submitted a paper 6 days ago

PEARL: Personalized Streaming Video Understanding Model

lyl010221-pku submitted a paper 12 days ago

Look Before Acting: Enhancing Vision Foundation Representations for Vision-Language-Action Models

View all activity

Papers

Know3D: Prompting 3D Generation with Knowledge from Vision-Language Models

PEARL: Personalized Streaming Video Understanding Model

View all Papers

PekingUniversity 's Papers 41

Submitted by

wenyue Chen

Know3D: Prompting 3D Generation with Knowledge from Vision-Language Models

PekingUniversity

Peking University

Submitted by

zhengyuanhong

PEARL: Personalized Streaming Video Understanding Model

PekingUniversity

Peking University

Submitted by

Zhifei Yang

FlowScene: Style-Consistent Indoor Scene Generation with Multimodal Graph Rectified Flow

PekingUniversity

Peking University

2

Submitted by

lin Luo

Look Before Acting: Enhancing Vision Foundation Representations for Vision-Language-Action Models

PekingUniversity

Peking University

3

Submitted by

Zeyu Zhang

MWM: Mobile World Models for Action-Conditioned Consistent Prediction

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

StereoAdapter-2: Globally Structure-Consistent Underwater Stereo Depth Estimation

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

MMA: Multimodal Memory Agent

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

MoRL: Reinforced Reasoning for Unified Motion Understanding and Generation

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

Light4D: Training-Free Extreme Viewpoint 4D Video Relighting

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

Code2Worlds: Empowering Coding LLMs for 4D World Generation

PekingUniversity

Peking University

Submitted by

Linli Yao

TimeChat-Captioner: Scripting Multi-Scene Videos with Time-Aware and Structural Audio-Visual Captions

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

GeneralVLA: Generalizable Vision-Language-Action Models with Knowledge-Guided Trajectory Planning

PekingUniversity

Peking University

Submitted by

ZehongMa

PixelGen: Pixel Diffusion Beats Latent Diffusion with Perceptual Loss

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

3D CoCa v2: Contrastive Learners with Test-Time Search for Generalizable Spatial Intelligence

PekingUniversity

Peking University

Submitted by

Jialong Wu

DocDancer: Towards Agentic Document-Grounded Information Seeking

PekingUniversity

Peking University

3

Submitted by

Zeyu Zhang

AnyDepth: Depth Estimation Made Easy

PekingUniversity

Peking University

Submitted by

Zhuofan Shi

MDAgent2: Large Language Model for Code Generation and Knowledge Q&A in Molecular Dynamics

PekingUniversity

Peking University

Submitted by

Zhicheng Sun

Learning from Next-Frame Prediction: Autoregressive Video Modeling Encodes Effective Representations

PekingUniversity

Peking University

Submitted by

bohan zeng

DataFlow: An LLM-Driven Framework for Unified Data Preparation and Workflow Automation in the Era of Data-Centric AI

PekingUniversity

Peking University

Submitted by

bohan zeng

Scone: Bridging Composition and Distinction in Subject-Driven Image Generation via Unified Understanding-Generation Modeling

PekingUniversity

Peking University

Submitted by

Yuyang Li

Simultaneous Tactile-Visual Perception for Learning Multimodal Robot Manipulation

PekingUniversity

Peking University

Submitted by

bohan zeng

VABench: A Comprehensive Benchmark for Audio-Video Generation

PekingUniversity

Peking University

Submitted by

YuchuanTian

From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

DragMesh: Interactive 3D Generation Made Easy

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

EgoLCD: Egocentric Video Generation with Long Context Diffusion

PekingUniversity

Peking University

Submitted by

Jianzong Wu

Does Hearing Help Seeing? Investigating Audio-Video Joint Denoising for Video Generation

PekingUniversity

Peking University

Submitted by

Weiyang Jin

Does Understanding Inform Generation in Unified Multimodal Models? From Analysis to Path Forward

PekingUniversity

Peking University

Submitted by

ZehongMa

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

PekingUniversity

Peking University

Submitted by

HuangTing

MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile Robots

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

EvoVLA: Self-Evolving Vision-Language-Action Model

PekingUniversity

Peking University

ID-Composer: Multi-Subject Video Synthesis with Hierarchical Identity Preservation

PekingUniversity

Peking University

Submitted by

kun ouyang

Conan: Progressive Learning to Reason Like a Detective over Multi-Scale Visual Evidence

PekingUniversity

Peking University

Submitted by

bohan zeng

Rethinking Driving World Model as Synthetic Data Generator for Perception Tasks

PekingUniversity

Peking University

Submitted by

Zongjian Li

Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedback

PekingUniversity

Peking University

Submitted by

Hu

Universal Image Restoration Pre-training via Masked Degradation Classification

PekingUniversity

Peking University

Submitted by

Zezhong Qian

WristWorld: Generating Wrist-Views via 4D World Models for Robotic Manipulation

PekingUniversity

Peking University

Can World Models Benefit VLMs for World Dynamics?

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

StereoAdapter: Adapting Stereo Depth Estimation to Underwater Scenes

PekingUniversity

Peking University

Submitted by

taesiri

Nav-R1: Reasoning and Navigation in Embodied Scenes

PekingUniversity

Peking University

Submitted by

Zeyu Zhang

3D-R1: Enhancing Reasoning in 3D VLMs for Unified Scene Understanding

PekingUniversity

Peking University

Submitted by

Feifan Song

TIME: A Multi-level Benchmark for Temporal Reasoning of LLMs in Real-World Scenarios

PekingUniversity

Peking University