Wenhu Chen

20

Papers

427

Total Citations

1

Affiliations

Affiliations

University of Waterloo

Papers (20)

VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning

UniIR: Training and Benchmarking Universal Multimodal Information Retrievers

General-Reasoner: Advancing LLM Reasoning Across All Domains

Reward Guided Latent Consistency Distillation

Harnessing Webpage UIs for Text-Rich Visual Understanding

VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation

Enhancing Human Experience in Human-Agent Collaboration: A Human-Centered Modeling Approach Based on Positive Human Gain

Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation

Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers

MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

Instruct-Imagen: Image Generation with Multi-modal Instruction

MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions

Video Captioning via Hierarchical Reinforcement Learning

Violin: A Large-Scale Dataset for Video-and-Language Inference

Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting

Local Explanation of Dialogue Response Generation

Counterfactual Maximum Likelihood Estimation for Training Deep Networks

Subject-driven Text-to-Image Generation via Apprenticeship Learning

MagicBrush: A Manually Annotated Dataset for Instruction-Guided Image Editing

MARBLE: Music Audio Representation Benchmark for Universal Evaluation