About Publications Experience Awards Talks CV


photo

weizhi.wang AT ucsb DOT edu

Weizhi Wang (王伟志)

weizhiwang AT ucsb DOT edu / LinkedIn / GitHub / Google Scholar / CV

I'm a second-year Ph.D. candidate advised by Prof. Xifeng Yan in University of California Santa Barbara. Before that, I obtained my master and bachelor degree from Rutgers University and Xi'an Jiaotong University, respectively.

Research Focus

I am focusing on the field of natural language processing. My research focus and past experience can be summarized as the following aspects:

  • Multimodal Large Language Models: Augmenting LLMs with visual-knowledge [VaLM]
  • Long-Context Large Language Models: Long-context LLMs with memory-cache augmentation [LongMem]
  • Data-Centric Research: Filtering and Demistifying the pre-training and instruction-tuning data for LLMs and Multimodal LLMs [MLM-Filter]

Additionally, I also worked on some application areas like:

Preprints

Finetuned Multimodal Language Models Are High-Quality Image-Text Data Filters
Weizhi Wang*, Khalil Mrini, Linjie Yang, Sateesh Kumar, Yu Tian, Xifeng Yan, Heng Wang
[arXiv] [project website] [code] [model] [data]

GPT-4V(ision) as a Generalist Evaluator for Vision-Language Tasks
Xinlu Zhang*, Yujie Lu*, Weizhi Wang*, An Yan, Jun Yan, Lianke Qin, Heng Wang, Xifeng Yan, William Y. Wang, Linda R. Petzold
[arXiv] (* Equal Contribution)

STEPS: A Benchmark for Order Reasoning in Sequential Tasks
Weizhi Wang, Hong Wang, Xifeng Yan
[arXiv] [code]

Bot or Human? Detecting ChatGPT Imposters with A Single Question
Hong Wang, Xuan Luo, Weizhi Wang, Xifeng Yan
[arXiv] [code]

Selected Publications

Augmenting Language Models with Long-Term Memory
Weizhi Wang, Li Dong, Hao Cheng, Xiaodong Liu, Xifeng Yan, Jianfeng Gao, Furu Wei.
NeurIPS 2023. [arXiv] [code]
Achievements:
1. Awarded as TOP 3 Popular ArXiv Papers on Twitter during June 2023
2. Ranked No. 15 on Github Trending on June 16th, 2023, 665 Stars in total
3. Media exposure AINLPer, WindowsReport, GenAI, Medium, ZDNet

GauchoChat: Towards Proactive, Controllable, and Personalized Social Conversation
Hong Wang, Weizhi Wang, Rajan Saini, Marina Zhukova, Xifeng Yan
Amazon Proceedings 2023. [Paper]
Champion🏆 for Amazon Alexa SocialBot Grand Challange 5.

Visually-Augmented Language Modeling
Weizhi Wang, Li Dong, Hao Cheng, Haoyu Song, Xiaodong Liu, Xifeng Yan, Jianfeng Gao, Furu Wei.
ICLR 2023. [arXiv] [code]

Non-Parametric Domain Adaptation for End-to-End Speech Translation
Yichao Du*, Weizhi Wang*, Zhirui Zhang, Boxing Chen, Tong xu, Jun Xie, Enhong Chen.
EMNLP 2022. Oral Presentation, Top 3%. (* Equal Contribution) [arXiv] [code]

Task-Oriented Dialogue Systems as Natural Language Generation
Weizhi Wang, Zhirui Zhang, Junliang Guo, Yinpei Dai, Boxing Chen, Weihua Luo.
SIGIR 2022. [arXiv] [code]

Regularizing End-to-End Speech Translation with Triangular Decomposition Agreement
Yichao Du, Zhirui Zhang, Weizhi Wang, Boxing Chen, Jun Xie, Tong Xu.
AAAI 2022. [arXiv] [code]

Rethinking Zero-shot Neural Machine Translation: From a Perspective of Latent Variables
Weizhi Wang, Zhirui Zhang, Yichao Du, Boxing Chen, Jun Xie, Weihua Luo.
EMNLP Findings 2021. [arXiv] [code]

Competition

The Fifth Amazon Alexa Prize SocialBot Grand Challenge. 2023.
Champion🏆, winning $250,000 first place bonus
As a member of GauchoChat, we developed a proactive, controllable, and personalized social chatbot. [Paper] [News]

Experience

Intelligent Creation Group, TikTok, San Jose, US. Sep 2023--Dec 2023
Research Intern
Mentor: Heng Wang

Natural Language Computing Group, Microsoft Research Asia, Beijing, CN. Nov 2021--May 2023
Research Intern, MSR Asia-MSR Redmond Joint StarLeap Program
Mentor: Li Dong, Hao Cheng, Xiaodong Liu, Furu Wei

Machine Translation Group, Alibaba DAMO Academy, Hangzhou, CN. Sep 2020--Nov 2021
Research Intern
Mentor: Zhirui Zhang and Boxing Chen

Awards

  • "Stars of Tomorrow" Excellent Intern Award, Microsoft Research, 2023.
  • Academic Excellence Fellowship. University of California Santa Barbara. 2022.
  • Outstanding Graduate Academic Performance Award. Rutgers University. 2021.

Talks

Revisiting GPT-Series and InstructGPT. Talk at UCSB CS291A. [Slides]

CLIP and VaLM. Talk at UCSB CS291A. [Slides]

Professional Services

- Conferences Reviewer: KDD 2023, NeurIPS 2023, ICLR 2024, SDM 2024.
- Journal Reviewer: TASLP, TALLIP.

Teaching

Rutgers CS170: Computer Application for Business, Teaching Assistant, Spring 2020 & Fall 2020 & Spring 2021. [Course Website]

UCSB CS165A: Artificial Intelligence, Teaching Assistant, Fall 2022. [Course Website]


modified from © Yangqing Jia 2013

Total Visits: