Zehan Wang

SpatialHand: Generative Object Manipulation from 3D Prespective

Depth Anything with Any Prior

Vox-Infinity: Benchmarking the Limits of Long-Context Spoken Language Models

FocusDiff: Advancing Fine-Grained Text-Image Alignment for Autoregressive Visual Generation through RL

AlignSep: Temporally-Aligned Video-Queried Sound Separation with Flow Matching

DSI-Bench: A Benchmark for Dynamic Spatial Intelligence

OmniChat: Enhancing Spoken Dialogue Systems with Scalable Synthetic Data for Diverse Scenarios

Orient Anything V2: Unifying Orientation and Rotation Understanding

VoxDialogue: Can Spoken Dialogue Systems Understand Information Beyond Words?

WavTokenizer: an Efficient Acoustic Discrete Codec Tokenizer for Audio Language Modeling

OmniBind: Large-scale Omni Multimodal Representation via Binding Spaces

OmniSep: Unified Omni-Modality Sound Separation with Query-Mixup

Diff-Prompt: Diffusion-Driven Prompt Generator with Mask Supervision

Improving Long-Text Alignment for Text-to-Image Diffusion Models

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models

ControlSpeech: Towards Simultaneous Zero-shot Speaker Cloning and Zero-shot Language Style Control

T2A-Feedback: Improving Basic Capabilities of Text-to-Audio Generation via Fine-grained AI Feedback

OmniChat: Enhancing Spoken Dialogue Systems with Scalable Synthetic Data for Diverse Scenarios

AVSET-10M: An Open Large-Scale Audio-Visual Dataset with High Correspondence

Noise-Robust Audio-Visual Speech-Driven Body Language Synthesis

Advancing Multimodal Unified Discrete Representations

Dynamic Switching Teacher: How to Generalize Temporal Action Detection Models

MindLoc: A Secure Brain-Based System for Object Localization