3 papers
cs.CL2024
Recurrent Alignment with Hard Attention for Hierarchical Text Rating
Chenxi Lin, Jiayu Ren, Guoxiu He +3
While large language models (LLMs) excel at understanding and generating plain text, they are not tailored to handle hierarchical text structures or directly predict task-specific…
cs.SI2024
Predicting Scientific Impact Through Diffusion, Conformity, and Contribution Disentanglement
Zhikai Xue, Guoxiu He, Zhuoren Jiang +4
The scientific impact of academic papers is influenced by intricate factors such as dynamic popularity and inherent contribution. Existing models typically rely on static graphs fo…
cs.CL2024
Evolving Knowledge Distillation with Large Language Models and Active Learning
Chengyuan Liu, Yangyang Kang, Fubang Zhao +4
Large language models (LLMs) have demonstrated remarkable capabilities across various NLP tasks. However, their computational costs are prohibitively high. To address this issue, p…