Min-Hung Chen

Research Scientist@NVIDIA·中国台湾·OpenReview

研究方向

Multi-modal Learning · Vision Transformer · Transfer Learning · Self-Supervised Learning · Domain Adaptation · Action Recognition · Video Understanding · Unsupervised Learning · Computer Vision · Deep Learning

Min-Hung Chen

FineBench: Benchmarking and Enhancing Vision-Language Models for Fine-grained Human Activity Understanding

HNDiff: Haze-Noise Diffusion for Image Dehazing

DLER: Doing Length pEnalty Right — Incentivizing More Intelligence per Token via Reinforcement Learning

MARS: Harmonizing Multimodal Convergence via Adaptive Rank Search

Zoom-Zero: Reinforced Coarse-to-Fine Video Understanding via Temporal Zoom-in

DDTNet: Degradation Disentanglement and Transfer Network for Domain-Adaptive All-in-One Image De-weathering

Learning Skills from Action-Free Videos

Hymba: A Hybrid-head Architecture for Small Language Models

SANER: Annotation-free Societal Attribute Neutralizer for Debiasing CLIP

ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning

BlurDM: A Blur Diffusion Model for Image Deblurring

Segment, Associate, and Classify: Decoupled Audio-Visual Segmentation Framework

HERMES: temporal-coHERent long-forM understanding with Episodes and Semantics

Temporal Prompting Matters: Rethinking Referring Video Object Segmentation

360-InpaintR: Reference-Guided 3D Inpainting for Unbounded Scenes