New Paradigm
"New paradigm" research encompasses diverse efforts to improve existing methods across various scientific fields. Current work focuses on developing novel algorithms and model architectures, such as those leveraging graph neural networks, generative adversarial networks, and large language models, to enhance efficiency, accuracy, and interpretability in tasks ranging from speech enhancement and image analysis to scientific discovery and recommendation systems. These advancements aim to address limitations in existing approaches, particularly concerning computational cost, data scarcity, and the explainability of complex models, ultimately impacting fields from materials science to healthcare. The overarching goal is to create more efficient, robust, and insightful tools for scientific investigation and practical applications.
Papers
Listen Again and Choose the Right Answer: A New Paradigm for Automatic Speech Recognition with Large Language Models
Yuchen Hu, Chen Chen, Chengwei Qin, Qiushi Zhu, Eng Siong Chng, Ruizhe Li
LLM and Simulation as Bilevel Optimizers: A New Paradigm to Advance Physical Scientific Discovery
Pingchuan Ma, Tsun-Hsuan Wang, Minghao Guo, Zhiqing Sun, Joshua B. Tenenbaum, Daniela Rus, Chuang Gan, Wojciech Matusik