Zobrazeno 1 - 10
of 11 493
pro vyhledávání: '"Bai, Yu"'
In Greek mythology, Pistis symbolized good faith, trust, and reliability. Drawing inspiration from these principles, Pistis-RAG is a scalable multi-stage framework designed to address the challenges of large-scale retrieval-augmented generation (RAG)
Externí odkaz:
http://arxiv.org/abs/2407.00072
Recent studies have demonstrated that In-Context Learning (ICL), through the use of specific demonstrations, can align Large Language Models (LLMs) with human preferences known as In-Context Alignment (ICA), indicating that models can comprehend huma
Externí odkaz:
http://arxiv.org/abs/2406.11474
Autor:
Bai, Yu, Zou, Xiyuan, Huang, Heyan, Chen, Sanxing, Rondeau, Marc-Antoine, Gao, Yang, Cheung, Jackie Chi Kit
Long sequence modeling has gained broad interest as large language models (LLMs) continue to advance. Recent research has identified that a large portion of hidden states within the key-value caches of Transformer models can be discarded (also termed
Externí odkaz:
http://arxiv.org/abs/2406.12018
Autor:
Wang, Tengbo, Bai, Yu
How to extract instance-level masks without instance-level supervision is the main challenge of weakly supervised instance segmentation (WSIS). Popular WSIS methods estimate a displacement field (DF) via learning inter-pixel relations and perform clu
Externí odkaz:
http://arxiv.org/abs/2406.18558
Autor:
Wei, Yukang, Bai, Yu
Temperature plays a pivotal role in moderating label softness in the realm of knowledge distillation (KD). Traditional approaches often employ a static temperature throughout the KD process, which fails to address the nuanced complexities of samples
Externí odkaz:
http://arxiv.org/abs/2404.12711
Light curves serve as a valuable source of information on stellar formation and evolution. With the rapid advancement of machine learning techniques, it can be effectively processed to extract astronomical patterns and information. In this study, we
Externí odkaz:
http://arxiv.org/abs/2404.10757
Large Language Models (LLMs) often memorize sensitive, private, or copyrighted data during pre-training. LLM unlearning aims to eliminate the influence of undesirable data from the pre-trained model while preserving the model's utilities on other tas
Externí odkaz:
http://arxiv.org/abs/2404.05868
Autor:
Luo, Changqing, Li, Jiao, Zheng, Chuanjie, Liu, Dongdong, Li, Zhenwei, Luo, Yangping, Nemeth, Peter, Zhang, Bo, Xiong, Jianping, Wang, Bo, Wang, Song, Bai, Yu, Li, Qingzheng, Wang, Pei, Han, Zhanwen, Liu, Jifeng, Huang, Yang, Chen, Xuefei, Liu, Chao
Although supernovae is a well-known endpoint of an accreting white dwarf, alternative theoretical possibilities has been discussing broadly, such as the accretion-induced collapse (AIC) event as the endpoint of oxygen-neon (ONe) white dwarfs, either
Externí odkaz:
http://arxiv.org/abs/2404.04835
Autor:
Wang, Shiyu, Feng, Yihao, Lan, Tian, Yu, Ning, Bai, Yu, Xu, Ran, Wang, Huan, Xiong, Caiming, Savarese, Silvio
Natural language serves as a common and straightforward control signal for humans to interact seamlessly with machines. Recognizing the importance of this interface, the machine learning community is investing considerable effort in generating data t
Externí odkaz:
http://arxiv.org/abs/2402.10941
Autor:
Bai, Yu, Huang, Heyan, Piano, Cesare Spinoso-Di, Rondeau, Marc-Antoine, Chen, Sanxing, Gao, Yang, Cheung, Jackie Chi Kit
In-context learning (ICL) has become an effective solution for few-shot learning in natural language processing. However, our understanding of ICL's working mechanisms is limited, specifically regarding how models learn to perform tasks from ICL demo
Externí odkaz:
http://arxiv.org/abs/2401.11323