Xin Dong

8

Papers

208

Total Citations

Papers (8)

ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models

Hymba: A Hybrid-head Architecture for Small Language Models

PAD: Patch-Agnostic Defense against Adversarial Patch Attacks

A Simple Background Augmentation Method for Object Detection with Diffusion Model

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Segment Every Out-of-Distribution Object

DreamFit: Garment-Centric Human Generation via a Lightweight Anything-Dressing Encoder

DOGR: Leveraging Document-Oriented Contrastive Learning in Generative Retrieval