Sentence representation learning is a crucial task in natural language processing, as the quality of learned representations directly influences downstream tasks, such as sentence …
G Qin, B Van Durme - International Conference on Machine …, 2023 - proceedings.mlr.press
Embedding text sequences is a widespread requirement in modern language understanding. Existing approaches focus largely on constant-size representations. This is …
Y Huang, P Zhao, Q Zhang, L Xing, H Wu, H Ma - Entropy, 2023 - mdpi.com
User alignment can associate multiple social network accounts of the same user. It has important research implications. However, the same user has various behaviors and friends …
S Wei, W Lu, X Peng, S Wang, YF Wang… - arXiv preprint arXiv …, 2023 - arxiv.org
By summarizing longer consumer health questions into shorter and essential ones, medical question answering (MQA) systems can more accurately understand consumer intentions …
Y Wu, X Pan, J Li, S Dou, J Dong… - International journal of …, 2024 - Wiley Online Library
Document representation is the basis of language modeling. Its goal is to turn natural language text that flows into a structured form that can be stored and processed by a …
H Wang, L Cheng, Z Li, DW Soh, L Bing - arXiv preprint arXiv:2310.10962, 2023 - arxiv.org
Contrastive learning has been proven to be effective in learning better sentence representations. However, to train a contrastive learning model, large numbers of labeled …
J Xu, W Shao, L Chen, L Liu - arXiv preprint arXiv:2305.13192, 2023 - arxiv.org
This paper improves contrastive learning for sentence embeddings from two perspectives: handling dropout noise and addressing feature corruption. Specifically, for the first …
H Wang, Z Li, L Cheng, L Bing - … of the 2024 Conference of the …, 2024 - aclanthology.org
Recently, large language models (LLMs) have emerged as a groundbreaking technology and their unparalleled text generation capabilities have sparked interest in their application …
The enhancement of unsupervised learning of sentence representations has been significantly achieved by the utility of contrastive learning. This approach clusters the …