Chunhui Zhang
Greetings! I am Chunhui Zhang (pronounced as Ch'un-hui Chang/張春晖), a Ph.D. student in Computer Science at Dartmouth🌲 , fortunately working with 🌟Professor Soroush Vosoughi .
I earned my MSCS degree (research-based) and was honored with the GSAS Fellowship at Brandeis University . My research journey began at Northeastern University , where I obtained a Bachelor's degree in CS and received the Outstanding Honor Thesis Award .
 / 
 / 
 / 
 / 
CV  / 
Google Scholar
Research
I am passionate about scaling modular cognitive architectures of foundation models, particularly supporting their memory capacity and long-range behaviors.
Posts
[Jan 22, 2025] Three papers are accepted to NAACL 2025. Thanks to my excellent collaborators and see you in Albuquerque!
[Sep 20, 2024] Two papers are accepted to EMNLP 2024. Thanks to my excellent collaborators and see you in Miami!
[Sep 7, 2024] Completed my summer internship at the Honda Research Institute in San Jose, CA. I am grateful for the collaborative experience with Dr. Shao-Yuan Lo and happy to have met great friends in the Bay Area.
Your browser does not support the video tag.
Overcoming Multi-step Complexity in Theory-of-Mind Reasoning: A Scalable Bayesian Planner
Chunhui Zhang , Sean Dae Houlihan, Kwonjoon Lee, Nakul Agarwal, Zhongyu Ouyang, {Soroush Vosoughi, Shao-Yuan Lo}
Preprint .
project page
/
preprint
Your browser does not support the video tag.
Superficial Self-Improved Reasoners Benefit from Model Merging
Xiangchi Yuan, Chunhui Zhang , Zheyuan Liu, Dachuan Shi, Soroush Vosoughi, Wenke Lee
Preprint .
project page
/
preprint
Your browser does not support the video tag.
Pretrained Image-Text Models are Secretly Video Captioners
Chunhui Zhang* , Yiren Jian*, Zhongyu Ouyang, Soroush Vosoughi
Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2025) Main Conference – Oral Presentation .
Using RL to post-train a Top-2 best multimodal captioner on PaperswithCode Leaderboard and beating industry-level captioners
project page
/
preprint
Your browser does not support the video tag.
Temporal Working Memory: Query-Guided Temporal Segment Refinement for Enhanced Multimodal Understanding
{Xingjian Diao*, Chunhui Zhang *}, Weiyi Wu, Zhongyu Ouyang, Peijun Qing, Ming Cheng, Soroush Vosoughi, Jiang Gui
Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2025) Findings.
Extending my working memory module study from EMNLP'24 to multimodal LLMs to enlarger their memory capabilities.
project page
/
preprint
Your browser does not support the video tag.
Is It Navajo? Accurate Language Detection in Endangered Athabaskan Languages
Ivory Yang, Weicheng Ma, Chunhui Zhang , Soroush Vosoughi
Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2025) Main Conference – Oral Presentation .
project page
/
preprint
Your browser does not support the video tag.
Working Memory Identifies Reasoning Limits in Language Models
Chunhui Zhang , Yiren Jian, Zhongyu Ouyang, Soroush Vosoughi
The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024 .
project page
/
preprint
Your browser does not support the video tag.
Expedited Training of Visual Conditioned Language Generation via Redundancy Reduction
Yiren Jian, Tingkai Liu, Yunzhe Tao, Chunhui Zhang , Soroush Vosoughi, Hongxia Yang
Annual Meeting of the Association for Computational Linguistics (ACL), 2024 .
project page
/
preprint
Your browser does not support the video tag.
Aligning Relational Learning with Lipschitz Fairness
{Yaning Jia, Chunhui Zhang }, Soroush Vosoughi
International Conference on Learning Representations (ICLR), 2024 .
project page
/
preprint
Your browser does not support the video tag.
Mitigating Emergent Robustness Degradation on Graphs while Scaling Up
{Xiangchi Yuan, Chunhui Zhang }, Yijun Tian, Yanfang Ye, et al.
International Conference on Learning Representations (ICLR), 2024 .
project page
/
preprint
Your browser does not support the video tag.
Symbol Prompt Tuning Completes the App Promotion Graph
Zhongyu Ouyang, Chunhui Zhang , Shifu Hou, Shang Ma, Chaoran Chen, Toby Li, Xusheng Xiao, Chuxu Zhang, Yanfang Ye
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2024 .
project page
/
preprint
Your browser does not support the video tag.
Graph Mixed Supervised Learning via Generalized Knowledge
Xiangchi Yuan, Yijun Tian, Chunhui Zhang , Yanfang Ye, Nitesh V Chawla, et al.
ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2024 .
project page
/
preprint
Your browser does not support the video tag.
GCVR: Reconstruction from Cross-View Enable Sufficient and Robust Graph Contrastive Learning
Qianlong Wen, Zhongyu Ouyang, Chunhui Zhang , Yiyue Qian, Chuxu Zhang, Yanfang Ye
The 40th Conference on Uncertainty in Artificial Intelligence (UAI), 2024 .
project page
/
preprint
Your browser does not support the video tag.
How to Improve Representation Alignment and Uniformity in Graph-based Collaborative Filtering?
Zhongyu Ouyang, Chunhui Zhang , Shifu Hou, Chuxu Zhang, Yanfang Ye
International AAAI Conference on Web and Social Media (ICWSM), 2024 .
project page
/
preprint
Your browser does not support the video tag.
Breaking the Trilemma of Privacy, Utility, and Efficiency via Controllable Machine Unlearning
{Zheyuan Liu, Guangyao Dou}, Yijun Tian, Chunhui Zhang , Eli Chien, Ziwei Zhu
ACM International World Wide Web Conference (The Web Conference/WWW), 2024 .
project page
/
preprint
Your browser does not support the video tag.
When Sparsity Meets Contrastive Models: Less Data Can Bring Better Class-Balanced Representations
Chunhui Zhang , Chao Huang, Yijun Tian, Qianlong Wen, Zhongyu Ouyang, Youhuan Li, Yanfang Ye, et al.
Fortieth International Conference on Machine Learning (ICML), 2023 .
project page
/
preprint
Your browser does not support the video tag.
Chasing All-Round Graph Representation Robustness: Model, Training, and Optimization
Chunhui Zhang , Yijun Tian, Mingxuan Ju, Zheyuan Liu, Yanfang Ye, Nitesh Chawla, et al.
The 11th International Conference on Learning Representations (ICLR) , 2023.
project page
/
preprint
Your browser does not support the video tag.
Mind the Gap: Mitigating the Distribution Gap in Graph Few-shot Learning
Chunhui Zhang , Hongfu Liu, Jundong Li, Yanfang Ye, et al.
Transactions on Machine Learning Research (TMLR) , 2023.
project page
/
preprint
Your browser does not support the video tag.
Fair Graph Representation Learning via Diverse Mixture-of-Experts
Zheyuan Liu*, Chunhui Zhang (Co-first author in alphabetical order)*, Yijun Tian, Erchi Zhang, Chao Huang, Yanfang Ye, et al.
International World Wide Web Conference (WWW) , 2023.
project page
/
preprint
Your browser does not support the video tag.
Boosting Graph Neural Networks via Adaptive Knowledge Distillation
Zhichun Guo, Chunhui Zhang , Yujie Fan, Yijun Tian, et al.
Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI) , 2023.
project page
/
preprint
Your browser does not support the video tag.
Heterogeneous Graph Masked Autoencoders
Yijun Tian, Kaiwen Dong, Chunhui Zhang , et al.
Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI) , 2023.
project page
/
preprint
Your browser does not support the video tag.
Heterogeneous Temporal Graph Neural Network Explainer
Jiazheng Li, Chunhui Zhang , et al.
ACM International Conference on Information and Knowledge Management (CIKM) , 2023.
project page
/
preprint
Your browser does not support the video tag.
Label-invariant Augmentation for Semi-Supervised Graph Classification
Han Yue, Chunhui Zhang , Chuxu Zhang, and Hongfu Liu
Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS) , 2022.
project page
/
preprint
Your browser does not support the video tag.
Co-Modality Imbalanced Graph Contrastive Learning
Yiyue Qian, Chunhui Zhang , Yiming Zhang, Qianlong Wen, Yanfang Ye, et al.
Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS) , 2022.
project page
/
preprint
Your browser does not support the video tag.
GraphBERT: Bridging Graph and Text for Malicious Behavior Detection on Social Media
Jiele Wu, Chunhui Zhang , Zheyuan Liu, Erchi Zhang, Steven Wilson, et al.
IEEE International Conference on Data Mining (ICDM) , 2022.
project page
/
preprint
Your browser does not support the video tag.
Look Twice as Much as You Say: Scene Graph Contrastive Learning for Self-Supervised Image Caption Generation
Chunhui Zhang , Chao Huang, Youhuan Li, Xiangliang Zhang, Yanfang Ye, et al.
ACM International Conference on Information and Knowledge Management (CIKM) , 2022.
project page
/
slides
/
poster
/
preprint
Your browser does not support the video tag.
Towards Tailored Models on Private AIoT Devices: Federated Direct Neural Architecture Search
Chunhui Zhang , Xiaoming Yuan, Qianyun Zhang, Guangxu Zhu, Lei Cheng, and Ning Zhang
IEEE Internet of Things Journal (IoTJ) , Feb. 2022.
project page
/
preprint
A fun fact:
Corresponding author Dr. Lei Cheng is the best professor I have ever met. As a fresh researcher, grateful for his very nice trust and professional advice.
To all students/friends
Feel free to drop me an email if you're up for a laid-back chat about life, career, or research. I'm dedicating (at-least) 30 minutes each week for these discussions, and I'm particularly eager to connect with students from underrepresented backgrounds or those dealing with challenges or inequity. Just reach out – I'm here to listen!
Fun Stuff
Racing - a happy part of my life. I particularly enjoy go-karting and circuit racing (some fun facts:
1st and 2nd place at Supercharged). But there is one type of racing that I have yet to try - my favorite rally driving (My favorite rally driver is Han Han).