Originally from Beijing, China, I am currently a Ph.D. student at MIT CSAIL advised by Yoon Kim. I work on natural language processing. I received my M.S. in computer science, B.S. in computer science, and B.A. in linguistics from the University of Washington.

I love computers, languages, and teaching computers to understand languages. On this path, I have done research at the Allen Institute for Artificial Intelligence (AI2) (as a PYI) and at the University of Washington with Prof. Noah Smith and Prof. Fei Xia. I have also worked on many interesting NLP and machine learning problems in the industry when I interned at Google, Meta, and other places.

With Alexis and Shannon, we have built a platform, cs-sop.org, where past applicants to CS PhD programs have generously shared their statements of purpose and other advice. We hope this could be a useful resource for future applicants.

• The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities
  Zhaofeng Wu, Xinyan Velocity Yu, Dani Yogatama, Jiasen Lu, and Yoon Kim
  To appear in International Conference on Learning Representations (ICLR), 2025.
• reWordBench: Benchmarking and Improving the Robustness of Reward Models with Transformed Inputs
  Zhaofeng Wu, Michihiro Yasunaga, Andrew Cohen, Yoon Kim, Asli Celikyilmaz, and Marjan Ghazvininejad
  Unpublished manuscript, 2025.
• SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models
  Yung-Sung Chuang, Benjamin Cohen-Wang, Shannon Zejiang Shen, Zhaofeng Wu, Hu Xu, Xi Victoria Lin, James Glass, Shang-Wen Li, and Wen-tau Yih
  Unpublished manuscript, 2025.
• Sparkle: Mastering Basic Spatial Capabilities in Vision Language Models Elicits Generalization to Composite Spatial Reasoning
  Yihong Tang, Ao Qu, Zhaokai Wang, Dingyi Zhuang, Zhaofeng Wu, Wei Ma, Shenhao Wang, Yunhan Zheng, Zhan Zhao, and Jinhua Zhao
  Unpublished manuscript, 2024.
• Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment
  Zhaofeng Wu, Ananth Balashankar, Yoon Kim, Jacob Eisenstein, and Ahmad Beirami
  In Empirical Methods in Natural Language Processing (EMNLP), 2024.
• Can You Learn Semantics Through Next-Word Prediction? The Case of Entailment
  William Merrill*, Zhaofeng Wu*, Norihito Naka, Yoon Kim, and Tal Linzen
  In Findings of the Annual Meeting of the Association for Computational Linguistics (ACL Findings), 2024.
• Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks
  Zhaofeng Wu, Linlu Qiu, Alexis Ross, Ekin Akyürek, Boyuan Chen, Bailin Wang, Najoung Kim, Jacob Andreas, and Yoon Kim
  In North American Chapter of the Association for Computational Linguistics (NAACL), 2024.
  [slides]
• ItiNera: Integrating Spatial Optimization with Large Language Models for Open-domain Urban Itinerary Planning
  Yihong Tang, Zhaokai Wang, Ao Qu, Yihao Yan, Zhaofeng Wu, Dingyi Zhuang, Jushi Kai, Kebing Hou, Xiaotong Guo, Jinhua Zhao, Zhan Zhao, and Wei Ma
  In Empirical Methods in Natural Language Processing (EMNLP): Industry Track, 2024.
• Transparency Helps Reveal When Language Models Learn Meaning
  Zhaofeng Wu, William Merrill, Hao Peng, Iz Beltagy, and Noah A. Smith
  In Transactions of the Association for Computational Linguistics (TACL), 2023.
  [slides] [poster]
• We're Afraid Language Models Aren't Modeling Ambiguity
  Alisa Liu, Zhaofeng Wu, Julian Michael, Alane Suhr, Peter West, Alexander Koller, Swabha Swayamdipta, Noah A. Smith, and Yejin Choi
  In Empirical Methods in Natural Language Processing (EMNLP), 2023.
• Continued Pretraining for Better Zero- and Few-Shot Promptability
  Zhaofeng Wu, Robert L. Logan IV, Pete Walsh, Akshita Bhagia, Dirk Groeneveld, Sameer Singh, and Iz Beltagy
  In Empirical Methods in Natural Language Processing (EMNLP), 2022.
  [poster]
• Modeling Context With Linear Attention for Scalable Document-Level Translation
  Zhaofeng Wu, Hao Peng, Nikolaos Pappas, and Noah A. Smith
  In Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP Findings), 2022.
  [poster]
• ABC: Attention with Bounded-memory Control
  Hao Peng, Jungo Kasai, Nikolaos Pappas, Dani Yogatama, Zhaofeng Wu, Lingpeng Kong, Roy Schwartz, and Noah A. Smith
  In Annual Meeting of the Association for Computational Linguistics (ACL), 2022.
• Learning with Latent Structures in Natural Language Processing: A Survey
  Zhaofeng Wu
  Unpublished manuscript, 2022.
• Infusing Finetuning with Semantic Dependencies
  Zhaofeng Wu, Hao Peng, and Noah A. Smith
  In Transactions of the Association for Computational Linguistics (TACL), 2021.
  [slides]
• Understanding Mention Detector-Linker Interaction in Neural Coreference Resolution
  Zhaofeng Wu and Matt Gardner
  In Workshop on Computational Models of Reference, Anaphora and Coreference @ EMNLP, 2021.
  [slides]
• Dynamic Sparsity Neural Networks for Automatic Speech Recognition
  Zhaofeng Wu, Ding Zhao, Qiao Liang, Jiahui Yu, Anmol Gulati, and Ruoming Pang
  In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021.
  [slides]
• WTMED at MEDIQA 2019: A Hybrid Approach to Biomedical Natural Language Inference
  Zhaofeng Wu, Yan Song, Sicong Huang, Yuanhe Tian, and Fei Xia
  In BioNLP Workshop and Shared Task @ ACL, 2019.

Talks

The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities
USC
November 2024

Can You Learn Semantics Through Next-Word Prediction? The Case of Entailment
Brown
April 2024

Generalization in the LLM Era
University of Utah, University of Virginia CS 6501
March 2024

Language Models: A Reality Check
UIUC CS 598, NYU, Cornell, Princeton, Google
September – November 2023

Graph Neural Networks for NLP
UW CSE 481N
May 2021

Awards

TIME Magazine’s Person of the Year in 2006
2008 感动中国年度人物特别奖