Zobrazeno 1 - 10
of 55
pro vyhledávání: '"Li, Liangyou"'
Autor:
Jiang, Yuxin, Huang, Bo, Wang, Yufei, Zeng, Xingshan, Li, Liangyou, Wang, Yasheng, Jiang, Xin, Shang, Lifeng, Tang, Ruiming, Wang, Wei
Direct preference optimization (DPO), a widely adopted offline preference optimization algorithm, aims to align large language models (LLMs) with human-desired behaviors using pairwise preference data. However, the winning response and the losing res
Externí odkaz:
http://arxiv.org/abs/2408.07471
Autor:
Wang, Zezhong, Zeng, Xingshan, Liu, Weiwen, Wang, Yufei, Li, Liangyou, Wang, Yasheng, Shang, Lifeng, Jiang, Xin, Liu, Qun, Wong, Kam-Fai
Current research found the issue of Early Answering in large language models (LLMs), where the models already have an answer before generating the Chain-of-Thought (CoT). This phenomenon suggests a potential lack of necessary dependency between the p
Externí odkaz:
http://arxiv.org/abs/2406.16144
Large language models (LLMs) have demonstrated remarkable performance on various natural language processing tasks. However, they are prone to generating fluent yet untruthful responses, known as "hallucinations". Hallucinations can lead to the sprea
Externí odkaz:
http://arxiv.org/abs/2406.11267
Autor:
Jiang, Yuxin, Wang, Yufei, Wu, Chuhan, Zhong, Wanjun, Zeng, Xingshan, Gao, Jiahui, Li, Liangyou, Jiang, Xin, Shang, Lifeng, Tang, Ruiming, Liu, Qun, Wang, Wei
Knowledge editing techniques, aiming to efficiently modify a minor proportion of knowledge in large language models (LLMs) without negatively impacting performance across other inputs, have garnered widespread attention. However, existing methods pre
Externí odkaz:
http://arxiv.org/abs/2402.11905
Autor:
Kwan, Wai-Chung, Zeng, Xingshan, Jiang, Yuxin, Wang, Yufei, Li, Liangyou, Shang, Lifeng, Jiang, Xin, Liu, Qun, Wong, Kam-Fai
Large language models (LLMs) are increasingly relied upon for complex multi-turn conversations across diverse real-world applications. However, existing benchmarks predominantly focus on single-turn evaluations, overlooking the models' capabilities i
Externí odkaz:
http://arxiv.org/abs/2401.16745
Autor:
Jiang, Yuxin, Wang, Yufei, Zeng, Xingshan, Zhong, Wanjun, Li, Liangyou, Mi, Fei, Shang, Lifeng, Jiang, Xin, Liu, Qun, Wang, Wei
The ability to follow instructions is crucial for Large Language Models (LLMs) to handle various real-world applications. Existing benchmarks primarily focus on evaluating pure response quality, rather than assessing whether the response follows cons
Externí odkaz:
http://arxiv.org/abs/2310.20410
Autor:
Kwan, Wai-Chung, Zeng, Xingshan, Wang, Yufei, Sun, Yusen, Li, Liangyou, Shang, Lifeng, Liu, Qun, Wong, Kam-Fai
Managing long sequences has become an important and necessary feature for large language models (LLMs). However, it is still an open question of how to comprehensively and systematically evaluate the long-sequence capability of LLMs. One of the reaso
Externí odkaz:
http://arxiv.org/abs/2310.19240
Autor:
Wang, Yufei, Zhong, Wanjun, Li, Liangyou, Mi, Fei, Zeng, Xingshan, Huang, Wenyong, Shang, Lifeng, Jiang, Xin, Liu, Qun
Large Language Models (LLMs) trained on extensive textual corpora have emerged as leading solutions for a broad array of Natural Language Processing (NLP) tasks. Despite their notable performance, these models are prone to certain limitations such as
Externí odkaz:
http://arxiv.org/abs/2307.12966
Length-controllable machine translation is a type of constrained translation. It aims to contain the original meaning as much as possible while controlling the length of the translation. We can use automatic summarization or machine translation evalu
Externí odkaz:
http://arxiv.org/abs/2305.02300
Utilizing pivot language effectively can significantly improve low-resource machine translation. Usually, the two translation models, source-pivot and pivot-target, are trained individually and do not utilize the limited (source, target) parallel dat
Externí odkaz:
http://arxiv.org/abs/2305.02261