Zhen Wang / 王震
Hi! I'm a computer science researcher working on natural language processing, machine learning, and data mining. I received my PhD from the Department of Computer Science and Engineering of The Ohio State University, advised by Prof. Huan Sun.
In summer 2022, I interned at MIT-IBM Watson AI Lab working with Rameswar, Yoon, Leonid, and Rogerio on efficient adaptation of large language models. In summer 2021, I was a research intern at the NLP group in Microsoft Research, Redmond, working with Nebojsa and Kolya from Mila, studying coherence boosting and prompting calibration on GPT-3. In summer 2020, I was a research intern at the Data Science team in NEC Laboratories, America working with Bo Zong, exploring commonsense knowledge representation and reasoning.
Email /
CV (Nov., 2022) /
GitHub /
Twitter /
Google Scholar /
LinkedIn /
Semantic Scholar
|
|
Research
I am interested in empowering current AI systems with more explicit and human-understandable knowledge, aiming to make them more generalizable, interpretable and data efficient. My research lies in the nexus of natural language processing, deep learning, data mining, and studies the "full stack" of the knowledge-centric AI from ground up: acquisition, representation, transfer, and reasoning. My long-term research goal is to transfuse strengths of human learning capabilities (e.g., intuitive physics, commonsense reasoning) to the next evolution of AI systems.
-
Knowledge Acquisition: Structured information extraction from text and graphs, knowledge graph construction, knowledge distillation from large language models
-
Knowledge Representation: Word representation learning, graph embedding learning, graph neural networks, commonsense concept learning
-
Knowledge Transfer: Transfer learning, multi-task learning, knowledge distillation, domain adaptation and generalization, few-shot learning
-
Knowledge Reasoning: Multi-hop reasoning over text and graphs (KG reasoning, complex QA), neuro-symbolic reasoning, commonsense reasoning
-
Applications: Natural language interfaces (dialogue systems, question answering), controllable text generation, text summarization, zero-/few-shot language model prompting, knowledge discovery for healthcare/bioinformatics
Free feel to reach out if you’d like to have a chat 🤗
|
News
-
01/2023: One paper, Entity Tracking via Effective Use of Multi-Task Learning Models, was accepted to EACL 2023!
-
01/2023: One paper, Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning, was accepted to ICLR 2023!
-
12/2022: Happy to finish my first in-person NLP course teaching and I'm impressed by what students have achieved after learning the very advanced NLP techniques. Check out this quick summary of their amazing final projects!
-
11/2022: Passed the PhD dissertation defense, Toward Knowledge-centric Natural Language Processing: Acquisition, Representation, Transfer, and Reasoning. Thanks to all my committee members, Prof. Huan Sun, Srinivasan Parthasarathy, Yu Su and Wei-Lun Chao.
-
08/2022: Will be teachinng CSE 5525: Foundations of Speech and Language Processing (Undergrad & Graduate) as an instructor at OSU this fall.
-
08/2022: Invited to serve as a PC member for AAAI 2023.
-
07/2022: Attended NAACL 2022 in Seattle. Presented our CQA Knowledge Transfer paper. Glad to meet all old and new friends!
-
06/2022: Invited to serve as a PC member for EMNLP 2022.
-
06/2022: Happy to give a talk about Efficient Adaptation of Large Language Models at the MIT Summer Working Group on Large LMs.
-
06/2022: Our TacoBot earned the third-place honor in the inaugural Alexa Prize TaskBot Challenge!
-
06/2022: Happy to give a tutorial about natural language processing and large language models in TDAI Deep Learning Summer School along with Prof. Huan Sun. Slides can be found here.
-
05/2022: Excited to join MIT-IBM Watson AI Lab in Cambridge, Boston as a research intern working with Rameswar Panda and Yoon Kim on efficient adaptation/pruning for large language models.
-
03/2022: Our team TacoBot moved forward to finals of the Alexa Prize TaskBot challenge!
-
03/2022: Honored to receive the 2022 Graduate Research Award of the CSE department.
-
02/2022: One paper, Coherence boosting: When your pretrained language model is not paying enough attention, was accepted to ACL 2022!
-
02/2022: Our team TacoBot moved forward to semifinals of the Alexa Prize TaskBot challenge! Try "Alexa, let's work together" or "Alexa, assist me" in your Alexa devices or the app when you want to do a DIY or cooking task!
-
12/2021: Passed the PhD candidacy exam with the proposal, "Knowledge-centric Natural Language Processing: Acquisition, Representation, and Reasoning". Thanks to all my committee members, Prof. Huan Sun, Srinivasan Parthasarathy, Yu Su and Wei-Lun Chao.
-
05/2021: Our team was selected to participate in the Alexa Prize TaskBot Challenge as one of 10 teams over 125 applications initiated from 15 countries! Looking forward to building a smart dialogue system to guide users through complex, multi-step plans (e.g., Cooking and DIY tasks) via multimodal interactions.
-
05/2021: Started Research Intern at Microsoft Research! Excited to work with Nebojsa Jojic and Kolya Malkin on Conditional Text Generation!
-
03/2021: Invited to serve as a PC member of EMNLP 2021.
-
03/2021: Attended WSDM 2021 virtually and present our work on modeling context pair interaction and learning graph pair emebddings.
-
03/2021: Honored to win Graduate Student Research Poster Award (Top 5) for 2021 Annual Student Research Poster Exhibition in our CSE department.
-
02/2021: Panelist on the panel discussion in Department of Astronomy, "2001: A Space Odyssey - Science Fiction vs Science Fact", discussing Artificial Intelligence, Life in Universe, Time & Relativity and Anthropology.
-
01/2021: Received SIGIR Student Travel Grant for WSDM 2021.
-
01/2021: Attended CDAC Rising Stars in Data Science Workshop with the agenda here.
-
12/2020: Honored to be selected to the Rising Stars in Data Science workshop hosted by the Center for Data and Computing (CDAC) at the University of Chicago.
-
12/2020: Invited to serve as a PC member of ACL 2021.
-
10/2020: Invited to serve as a PC member of NAACL 2021.
-
10/2020: One paper, Modeling Context Pair Interaction for Pairwise Tasks on Graphs, was accepted to WSDM 2021 (Acceptance Rate: 18.6%).
-
07/2020: Attended ACL 2020 virtually and presented our paper via QA sessions and pre-recorded video.
-
05/2020: Started Research Intern at NEC Labs America! Excited to work with Dr. Bo Zong on Commonsense Reasoning for NLU.
-
04/2020: One paper, Rationalizing Medical Relation Prediction from Corpus-level Statistics, about building self-interpretable deep learning model for relation prediction was accepted by ACL 2020 (Acceptance Rate: 22.7%).
-
04/2020: Invited to serve as a PC member of NLPCC 2020.
-
09/2019: One paper, Graph Embedding on Biomedical Networks: Methods, Applications, and Evaluations, was accepted by Bioinformatics.
-
08/2019: Attended KDD 2019 in Anchorage, Alaska and presented our work in an oral talk.
-
06/2019: Received SIGKDD Student Travel Award for KDD 2019.
-
04/2019: One paper, SurfCon: Synonym Discovery on Privacy-Aware Clinical Data, about knowledge extraction from text was accepted by KDD 2019 (Research Track, Acceptance Rate: 14.2%, Oral).
-
08/2018: One paper about Code Summarization was accepted by KDD 2018, Deep Learning Day.
-
06/2018: Attended NAACL 2018 in New Orleans, LA.
|
|
Batu Ozturkler, Nikolay Malkin, Zhen Wang, Nebojsa Jojic
[arXiv] 2210.01293
|
|
Zhen Wang, Rameswar Panda, Leonid Karlinsky, Rogerio Feris, Huan Sun, Yoon Kim
[ICLR 2023] The Eleventh International Conference on Learning Representations
PDF / Code / Slides / Poster
We propose Multitask Prompt Tuning (MPT) to exploit the rich cross-task knowledge for more efficient and generalizable transfer learning. MPT learns a single trasnferrable soft prompt through the use of a novel combination of prompt decomposition and prompt distillation.
|
|
Janvijay Singh, Fan Bai, Zhen Wang
[EACL 2023] The 17th Conference of the European Chapter of the Association for Computational Linguistics (Main)
PDF / Code / Slides / Poster
How to transfer multi-task knowledge from pre-training to niche downstream tasks, such as entity tracking on the procedural text? We show that you can reach STOA performance by simply fine-tuning T5 but with specialized QA prompt and task-specific decoding.
|
|
Nikolay Malkin, Zhen Wang, Nebojsa Jojic
[ACL 2022] The 60th Annual Meeting of the Association for Computational Linguistics
PDF / Code / Slides / Poster (Long Paper, Oral Presentation)
We demonstrate that large language models have insufficiently learned the effect of distant words on next-token prediction. We present Coherence Boosting, an inference procedure that increases a LM’s focus on a long context, which gets huge improvement on NLG and NLU tasks.
|
|
Lingbo Mo*, Zhen Wang*, Jie Zhao, Huan Sun
[SUKI@NAACL 2022] NAACL 2022 Structured and Unstructured Knowledge Integration
PDF / Code / Slides / Poster *Equal contribution
We study knowledge transfer for multi-hop reasoning processes between structured (Knowledge Base) and unstructred (text corpus) knowledge. We design SimultQA unifying KBQA and TextQA systems and leverage it to study how the reasoning is transferred between two knowledge sources.
|
|
Shijie Chen, Ziru Chen, Xiang Deng, Ashley Lewis, Lingbo Mo, Samuel Stevens, Zhen Wang, Xiang Yue, Tianshu Zhang, Yu Su, Huan Sun
[Alexa Prize TaskBot Challenge] 1st Proceedings of Alexa Prize TaskBot (Alexa Prize 2021)
PDF / Third-place honor in the TaskBot Finals!
We build TacoBot, a task-oriented dialogue system for the inaugural Alexa Prize TaskBot Challenge to assist users in multi-step cooking and home improvement tasks. We propose several data augmentation methods, such as GPT-3 simulation to bootstrap neural dialogue systems into new domains and make them more robust to noise user initiatives.
|
|
Zhen Wang, Bo Zong, Huan Sun
[WSDM 2021] The 14th ACM International Conference on Web Search and Data Mining
PDF / Code / Slides / Poster (Long Paper, Online Presentation)
We propose to explicitly model context interactions for pairwise prediction tasks on graphs, which consists of two perspectives, node-centric and pair-centric. We also propose to pre-train pair embeddings to facilitate the pair-centric model.
|
|
Zhen Wang, Jennifer Lee, Simon Lin, Huan Sun
[ACL 2020] The 58th Annual Meeting of the Association for Computational Linguistics
PDF / Code / Slides / Poster / Video (Long Paper, Online Presentation)
We propose a self-interpretable framework to rationalize the neural relation prediction based on corpus-level statistics. This framework is inspired by human cognitive theory about recall and recognition, which provides structured knowledge triplets as rationales.
|
|
Xiang Yue, Zhen Wang, Jingong Huang, Srinivasan Parthasarathy, Soheil Moosavinasab, Yungui Huang, Simon Lin, Wen Zhang, Ping Zhang, Huan Sun
[Bioinformatics] Volume 36, Issue 4, 15 February 2020, Pages 1241-1251
PDF / Code / Slides / Poster
We benchmark 11 representative graph embedding methods on 5 important biomedical tasks. We verify the effectivenes of recent graph embedding methods and provide general guidelines for their usage.
|
|
Zhen Wang, Xiang Yue, Soheil Moosavinasab, Yungui Huang, Simon Lin, Huan Sun
[KDD 2019] The 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PDF / Code / Slides / Poster (Research Track, Long Paper, Oral Presentation)
We propose to discover structured knowledge, synonyms from privacy-aware text corpus and present a novel framework to leverage both surface form and context information to discover out-of-distribution synonyms.
|
|
Jayavardhan Reddy Peddamail, Ziyu Yao, Zhen Wang, Huan Sun
[KDD 2018 Deep Learning Day] The 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PDF / Code / Slides / Poster (SPOTLIGHT)
We examine three popular datasets mined from Stack Overflow on the code summarization task and show that StaQC (Stack Overflow Question-Code pairs) helps achieve substantially better results.
|
|
Weifeng Liu, Zhen Wang, Dapeng Tao, Jun Yu
[MMM 2015] The 21th International Conference on Multimedia Modeling
PDF / Code / Slides / Poster /
Bibtex
@inproceedings{liu2015hessian,
title={Hessian regularized sparse coding for human action recognition},
author={Liu, Weifeng and Wang, Zhen and Tao, Dapeng and Yu, Jun},
booktitle={International Conference on Multimedia Modeling},
pages={502--511},
year={2015},
organization={Springer}
}
We propose Hessian regularized sparse coding (HessianSC) for action recognition, which can well preserve the local geometry and steer the sparse coding varying linearly along the manifold of data distribution.
|
Honors and Awards
-
Third-Place Honor, Inaugural Alexa Prize TaskBot Challenge, 2022
-
Graduate Research Award, CSE, OSU, 2022
-
Graduate Student Research Poster Award (Top 5), CSE, OSU, 2021
-
SIGIR Student Travel Grant, 2021
-
Rising Stars in Data Science, Center for Data and Computing (CDAC), University of Chicago, January 2021
-
SIGKDD Student Travel Award, 2019
-
China Scholarship Council (CSC) Scholarship for a fully funded visiting program in Polytech Nice Sophia, Nice, France, 2015
-
National Scholarship, China, 2014
-
Soong Ching Ling Foundation (SCLF) Scholarship, China, 2013
-
National Scholarship for Encouragement, China, 2012
|
Services
Program Committee Member:
-
ACL ARR (Oct'21, Nov'21, Jan'22, Apr'22, Sep'22, Oct'22, Dec'22)
-
SUKI 2022 Workshop at NAACL 2022
-
EMNLP 2021, 2022
-
ACL 2021, 2023
-
NAACL 2021
-
KDD 2023
-
AAAI 2023
-
NLPCC (2020, 2021, 2022)
External Reviewer:
-
KDD (2019, 2020), ACL 2018, ICDM 2018
|
|