Research
My research interests mainly lie in Natural Language Processing, specifically in leveraging machines as guides to comprehend the human world, exploring the reasoning abilities of LLMs, commonsense reasoning, and NLP for social good.
|
|
Knowledge Crosswords: Geometric Reasoning over Structured Knowledge with Large Language Models
Wenxuan Ding*, Shangbin Feng*, Yuhan Liu, Zhaoxuan Tan, Vidhisha Balachandran, Tianxing He, Yulia Tsvetkov
arXiv, 2023.
code
We propose KnowledgeCrosswords, a benchmark evaluating LLMs' abilities for geometric knowledge reasoning, posing new challenges involving reasoning with uncertainty, verification, backtracking, and more.
|
|
CAR: Conceptualization-Augmented Reasoner for Zero-Shot Commonsense Question Answering
Weiqi Wang*, Tianqing Fang*, Wenxuan Ding, Baixuan Xu, Xin Liu, Yangqiu Song, Antoine Bosselut
Fingdings of EMNLP, 2023.
code
|
|
QADYNAMICS: Training Dynamics-Driven Synthetic QA Diagnostic for Zero-Shot Commonsense Question Answering
Haochen Shi*, Weiqi Wang*, Tianqing Fang, Baixuan Xu, Wenxuan Ding, Xin Liu, Yangqiu Song
Fingdings of EMNLP, 2023.
code
|
|