Hui Zhang
6 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
The paper introduces SURE, a unified framework designed to standardize and improve the comparability and reproducibility of evaluations for advanced speech understanding models.
The paper introduces SURE, a unified framework designed to standardize and improve the comparability and reproducibility of evaluations for advanced speech understanding models.
This paper proposes a universal PET image denoising network (UniPET) to adaptively recover different dose reduction factor styles and achieve high-quality PET image denoising across various dose reduction factors.
This paper documents and analyzes the failure process of strategies used to address conceptual drift in long-horizon LLM collaboration and introduces the concept of 'Index Sickness' and the 'Pang Principle'. It also presents a physical engineering mechanism called 'Baseline-Log Physical Separation' to address this issue.
JoyNexus is a multi-tenant service for VLA model supervised fine-tuning, reinforcement learning, and evaluation, which decouples services, introduces group batching, and improves training efficiency.
The paper proposes a new method, AIMS, to improve instruction-guided generative recommendation by converting the effect of removing individual history events into ranking supervision.
Papers
When History Misleads: Asymmetric Margin Supervision for Instruction-Guided LLM Generative Recommendation
Ming Yin, Yuhan Yang, Chen Chen, Xinyu Lin +9 more
The paper proposes a new method, AIMS, to improve instruction-guided generative recommendation by converting the effect of removing individual history events into ranking supervision.