Zobrazeno 1 - 10
of 52
pro vyhledávání: '"Dai, Xinyu"'
Large language models demonstrate reasonable multilingual abilities, despite predominantly English-centric pretraining. However, the spontaneous multilingual alignment in these models is shown to be weak, leading to unsatisfactory cross-lingual trans
Externí odkaz:
http://arxiv.org/abs/2407.16222
Autor:
Wang, Yidong, Guo, Qi, Yao, Wenjin, Zhang, Hongbo, Zhang, Xin, Wu, Zhen, Zhang, Meishan, Dai, Xinyu, Zhang, Min, Wen, Qingsong, Ye, Wei, Zhang, Shikun, Zhang, Yue
This paper introduces AutoSurvey, a speedy and well-organized methodology for automating the creation of comprehensive literature surveys in rapidly evolving fields like artificial intelligence. Traditional survey paper creation faces challenges due
Externí odkaz:
http://arxiv.org/abs/2406.10252
Large language models (LLMs) exhibit robust capabilities in text generation and comprehension, mimicking human behavior and exhibiting synthetic personalities. However, some LLMs have displayed offensive personality, propagating toxic discourse. Exis
Externí odkaz:
http://arxiv.org/abs/2406.04583
Multimodal Large Language Models (MLLMs) are widely regarded as crucial in the exploration of Artificial General Intelligence (AGI). The core of MLLMs lies in their capability to achieve cross-modal alignment. To attain this goal, current MLLMs typic
Externí odkaz:
http://arxiv.org/abs/2405.14129
Emotion Recognition in Conversation (ERC) involves detecting the underlying emotion behind each utterance within a conversation. Effectively generating representations for utterances remains a significant challenge in this task. Recent works propose
Externí odkaz:
http://arxiv.org/abs/2403.20289
Autor:
Gao, Yuan, Zhu, Yiheng, Cao, Yuanbin, Zhou, Yinzhi, Wu, Zhen, Chen, Yujie, Wu, Shenglan, Hu, Haoyuan, Dai, Xinyu
Open Domain Multi-Hop Question Answering (ODMHQA) plays a crucial role in Natural Language Processing (NLP) by aiming to answer complex questions through multi-step reasoning over retrieved information from external knowledge sources. Recently, Large
Externí odkaz:
http://arxiv.org/abs/2403.12393
Autor:
Xing, Shangyu, Zhao, Fei, Wu, Zhen, An, Tuo, Chen, Weihao, Li, Chunhui, Zhang, Jianbing, Dai, Xinyu
Multimodal large language models (MLLMs) have attracted increasing attention in the past few years, but they may still generate descriptions that include objects not present in the corresponding images, a phenomenon known as object hallucination. To
Externí odkaz:
http://arxiv.org/abs/2402.09801
Large language models have shown impressive capabilities across a variety of NLP tasks, yet their generating text autoregressively is time-consuming. One way to speed them up is speculative decoding, which generates candidate segments (a sequence of
Externí odkaz:
http://arxiv.org/abs/2401.06706
Publikováno v:
2023 International Joint Conference on Neural Networks (IJCNN), Gold Coast, Australia, 2023, pp. 1-9
BERT-based models have shown a remarkable ability in the Chinese Spelling Check (CSC) task recently. However, traditional BERT-based methods still suffer from two limitations. First, although previous works have identified that explicit prior knowled
Externí odkaz:
http://arxiv.org/abs/2312.16623
Canonical relation extraction aims to extract relational triples from sentences, where the triple elements (entity pairs and their relationship) are mapped to the knowledge base. Recently, methods based on the encoder-decoder architecture are propose
Externí odkaz:
http://arxiv.org/abs/2312.07088