Thrilled to share some life updates! I've officially completed my Ph.D. journey at
@LTIatCMU
and started an exciting new role as a Senior Researcher at
@Microsoft
!
New preprint! DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Bases. Our proposed method achieved new SoTAs on KBQA! (1/5)
Arxiv link:
New preprint! DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Bases. Our proposed method achieved new SoTAs on KBQA! (1/5)
Arxiv link:
Our paper "Knowledge Embedding Based Graph Convolutional Network" has been accepted to
#www2021
Want to incorporate KE methods like TransH, TransD, QuatE... into GCN? Check an old version of this paper: . New version will be updated soon.
#StopAsianHate
CMU Graduate Student Assembly: Petition of Asian And Asian American students to the CMU GSA executive committee - Sign the Petition! via
@Change
I am excited to share our paper “Dr.Spider: A Diagnostic Evaluation Benchmark towards Text-to-SQL Robustness” got accepted as a top 5% paper (oral) to
#ICLR2023
.
We propose a novel framework that jointly generates both logical forms and direct answers, and then combines them for the final answers. Moreover, DecAF is based on simple free-text retrieval without relying on any entity linking tools. (3/5)
Previous methods either generate logical forms that can be executed over KBs to obtain final answers or predict answers directly. Empirical results show that the former often produces more accurate answers, but it suffers from non-execution issues. (2/5)
Our model achieves new state-of-the-art results on WebQSP, GrailQA, and FreebaseQA benchmarks, and gets very competitive results on the ComplexWebQuestions benchmark. (4/5)
Please sign and share this petition to reinstate UIUC graduate worker Ivor Chen after the University decided to dismiss him for behavior that they admit deserves a testing exemption! This is cruel and unjust, and we must stand up to UIUC on his behalf!
Together with Sheng Zhang, Patrick Ng, Henghui Zhu, Alexander Hanbo Li, Jun Wang, Yiqun Hu,
@WilliamWangNLP
, Zhiguo Wang, Bing Xiang, in collaboration with AWS AI Labs. (5/5)
We often cite papers using arXiv info w/o noting that they are already PUBLISHED in confs like
@aclmeeting
and
@iclr_conf
. These incorrect bib entries are so annoying!😠 We introduce Rebiber, a tool to fix them using
@aclanthology
and
@dblp_org
. 😆
Code: