Zobrazeno 1 - 10
of 635
pro vyhledávání: '"Sun Maosong"'
Autor:
Xiao, Chaojun, Cai, Jie, Zhao, Weilin, Zeng, Guoyang, Lin, Biyuan, Zhou, Jie, Zheng, Zhi, Han, Xu, Liu, Zhiyuan, Sun, Maosong
Large Language Models (LLMs) have emerged as a milestone in artificial intelligence, and their performance can improve as the model size increases. However, this scaling brings great challenges to training and inference efficiency, particularly for d
Externí odkaz:
http://arxiv.org/abs/2412.04315
Semantic map models (SMMs) construct a network-like conceptual space from cross-linguistic instances or forms, based on the connectivity hypothesis. This approach has been widely used to represent similarity and entailment relationships in cross-ling
Externí odkaz:
http://arxiv.org/abs/2412.01423
Autor:
Zeng, Zheni, Chen, Yuxuan, Yu, Shi, Yan, Yukun, Liu, Zhenghao, Wang, Shuo, Han, Xu, Liu, Zhiyuan, Sun, Maosong
Humans can utilize techniques to quickly acquire knowledge from specific materials in advance, such as creating self-assessment questions, enabling us to achieving related tasks more efficiently. In contrast, large language models (LLMs) usually reli
Externí odkaz:
http://arxiv.org/abs/2411.14790
Autor:
Zhao, Haozhe, Si, Shuzheng, Chen, Liang, Zhang, Yichi, Sun, Maosong, Zhang, Mingjia, Chang, Baobao
Large vision-language models (LVLMs) have achieved impressive results in various vision-language tasks. However, despite showing promising performance, LVLMs suffer from hallucinations caused by language bias, leading to diminished focus on images an
Externí odkaz:
http://arxiv.org/abs/2411.14279
Autor:
Fan, Shengda, Cong, Xin, Fu, Yuepeng, Zhang, Zhong, Zhang, Shuyan, Liu, Yuanwei, Wu, Yesai, Lin, Yankai, Liu, Zhiyuan, Sun, Maosong
Recent advancements in large language models (LLMs) have driven a revolutionary paradigm shift in process automation from Robotic Process Automation to Agentic Process Automation by automating the workflow orchestration procedure based on LLMs. Howev
Externí odkaz:
http://arxiv.org/abs/2411.05451
Autor:
Lin, Junming, Fang, Zheng, Chen, Chi, Wan, Zihao, Luo, Fuwen, Li, Peng, Liu, Yang, Sun, Maosong
The rapid development of Multimodal Large Language Models (MLLMs) has expanded their capabilities from image comprehension to video understanding. However, most of these MLLMs focus primarily on offline video comprehension, necessitating extensive pr
Externí odkaz:
http://arxiv.org/abs/2411.03628
Activation sparsity denotes the existence of substantial weakly-contributed elements within activation outputs that can be eliminated, benefiting many important applications concerned with large language models (LLMs). Although promoting greater acti
Externí odkaz:
http://arxiv.org/abs/2411.02335
This study explores the tokenization of multitrack sheet music in ABC notation, introducing two methods--bar-stream and line-stream patching. We compare these methods against existing techniques, including bar patching, byte patching, and Byte Pair E
Externí odkaz:
http://arxiv.org/abs/2410.17584
Autor:
Si, Shuzheng, Zhao, Haozhe, Chen, Gang, Li, Yunshui, Luo, Kangyang, Lv, Chuancheng, An, Kaikai, Qi, Fanchao, Chang, Baobao, Sun, Maosong
The expansion of large language models to effectively handle instructions with extremely long contexts has yet to be fully investigated. The primary obstacle lies in constructing a high-quality long instruction-following dataset devised for long cont
Externí odkaz:
http://arxiv.org/abs/2410.15633
Autor:
Li, Xinze, Mei, Sen, Liu, Zhenghao, Yan, Yukun, Wang, Shuo, Yu, Shi, Zeng, Zheni, Chen, Hao, Yu, Ge, Liu, Zhiyuan, Sun, Maosong, Xiong, Chenyan
Retrieval-Augmented Generation (RAG) has proven its effectiveness in mitigating hallucinations in Large Language Models (LLMs) by retrieving knowledge from external resources. To adapt LLMs for RAG pipelines, current approaches use instruction tuning
Externí odkaz:
http://arxiv.org/abs/2410.13509