Výsledky vyhledávání

Report

Your Diffusion Model is Secretly a Noise Classifier and Benefits from Contrastive Training

Autor: Wu, Yunshu, Luo, Yingtao, Kong, Xianghao, Papalexakis, Evangelos E., Steeg, Greg Ver

Diffusion models learn to denoise data and the trained denoiser is then used to generate new samples from the data distribution. In this paper, we revisit the diffusion sampling process and identify a fundamental cause of sample quality degradation:

Externí odkaz: http://arxiv.org/abs/2407.08946

Zobrazit plný text záznamu

Report

Less is More for Long Document Summary Evaluation by LLMs

Autor: Wu, Yunshu, Iso, Hayate, Pezeshkpour, Pouya, Bhutani, Nikita, Hruschka, Estevam

Large Language Models (LLMs) have shown promising performance in summary evaluation tasks, yet they face challenges such as high computational costs and the Lost-in-the-Middle problem where important information in the middle of long documents is oft

Externí odkaz: http://arxiv.org/abs/2309.07382

Zobrazit plný text záznamu

Report

High-Performance and Flexible Parallel Algorithms for Semisort and Related Problems

Autor: Dong, Xiaojun, Wu, Yunshu, Wang, Zhongqi, Dhulipala, Laxman, Gu, Yan, Sun, Yihan

Semisort is a fundamental algorithmic primitive widely used in the design and analysis of efficient parallel algorithms. It takes input as an array of records and a function extracting a \emph{key} per record, and reorders them so that records with e

Externí odkaz: http://arxiv.org/abs/2304.10078

Zobrazit plný text záznamu