Alyssa, Yi CHENG (程轶)

Ph.D. Student

Department of Computing,
The Hong Kong Polytechnic University,
Hung Hom, Kowloon, Hong Kong, China

Resume | Email | Google Scholar

I am a fourth-year Ph.D. candidate at The Hong Kong Polytechnic University, specializing at Natural Language Processing (NLP) under the mentorship of Prof. Maggie Wenjie Li. I earned my Bachelor's degree in Computer Science from Peking University, where I had the privilege of working with Prof. Sujian Li. Currently, I am engaged as a research intern at Microsoft Research Asia, where I am honored to work with Dr. Yeyun Gong and Dr. Xiao Liu. Before that, I also had the opportunity to intern at Tencent Jarvis Research Center, where I contributed under the guidance of Prof. Yefeng Zheng and Prof. Bang Liu.

Research Interest

My research interest broadly lies in the NLP techniques for social good. In recent years, my research has specifically focused on the following directions:

NLP applications for mental healthcare (e.g. emotional support conversation systems, empathetic chatbots).
NLP applications in the medical field, especially medical consultation dialogue systems.

Previously, I also did some research on Chinese discourse parsing and question generation in my undergraduate years.

Preprint

Integrative Decoding: Improve Factuality via Implicit Self-consistency [arXiv] [code]
Yi Cheng, Xiao Liang, Yeyun Gong, Wen Xiao, Song Wang, Yuji Zhang, Wenjun Hou, Kaishuai Xu, Wenge Liu, Wenjie Li, Jian Jiao, Qi Chen, Peng Cheng, Wayne Xiong
AutoPal: Autonomous Adaptation to Users for Personal AI Companionship [arXiv]
Yi Cheng, Wenge Liu, Kaishuai Xu, Wenjun Hou, Yi Ouyang, Chak Tou Leong, Xian Wu, Yefeng Zheng
No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks [arXiv]
Chak Tou Leong, Yi Cheng, Kaishuai Xu, Jian Wang, Hanlin Wang, Wenjie Li

Publications

COOPER: Coordinating Specialized Agents towards a Complex Dialogue Goal [arXiv] [code]
Yi Cheng, Wenge Liu, Jian Wang, Chak Tou Leong, Yi Ouyang, Wenjie Li, Xian Wu, Yefeng Zheng
AAAI 2024.
Self-Detoxifying Language Models via Toxification Reversal [arXiv] [code]
Chak Tou Leong*, Yi Cheng*, Jiashuo Wang, Jian Wang, and Wenjie Li. (*: Equal Contribution)
EMNLP 2023.
ICON: Improving Inter-Report Consistency of Radiology Report Generation via Lesion-aware Mix-up Augmentation. [arXiv]
Wenjun Hou, Yi Cheng*, Kaishuai Xu, Yan Hu, Wenjie Li, Jiang Liu.
EMNLP 2024 Findings.
Reasoning Like a Doctor: Improving Medical Dialogue Systems via Diagnostic Reasoning Process Alignment [arXiv]
Kaishuai Xu, Yi Cheng*, Wenjun Hou*, Qiaoyu Tan, Wenjie Li.
ACL 2024 Findings.
Target-oriented Proactive Dialogue Systems with Personalization: Problem Formulation and Dataset Curation [arXiv] [code]
Jian Wang, Yi Cheng, Dongding Lin, Chak Tou Leong, and Wenjie Li.
EMNLP 2023.
RECAP: Towards Precise Radiology Report Generation via Dynamic Disease Progression Reasoning [arXiv] [code]
Wenjun Hou, Yi Cheng*, Kaishuai Xu*, Wenjie Li, Jiang Liu. (*: Equal Contribution)
EMNLP Findings, 2023.
ORGan: Observation-Guided Radiology Report Generation via Tree Reasoning [arXiv] [code]
Wenjun Hou, Yi Cheng*, Kaishuai Xu*, Wenjie Li, Jiang Liu. (*: Equal Contribution)
ACL 2023.
Medical Dialogue Generation via Dual Flow Modeling [arXiv]
Kaishuai Xu, Wenjun Hou*, Yi Cheng*, Jian Wang, Wenjie Li. (*: Equal Contribution)
ACL Findings, 2023.
Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning [arXiv] [code]
Yi Cheng*, Wenge Liu*, Wenjie Li, Jiashuo Wang, Ruihui Zhao, Bang Liu, Xiaodan Liang, Yefeng Zheng. (*: Equal Contribution)
EMNLP 2022
CARE: Causality Reasoning for Empathetic Responses by Conditional Graph Generation [arXiv] [code]
Jiashuo Wang, Yi Cheng, Wenjie Li.
EMNLP Findings, 2022.
MedDG: An Entity-Centric Medical Consultation Dataset for Entity-Aware Medical Dialogue Generation [arXiv] [code]
Wenge Liu*, Jianheng Tang*, Yi Cheng*, Wenjie Li, Yefeng Zheng, Xiaodan Liang. (*: Equal Contribution)
NLPCC 2022.
“My nose is running.”“Are you also coughing?”: Building A Medical Diagnosis Agent with Interpretable Inquiry Logics [arXiv] [code]
Wenge Liu*, Yi Cheng*, Hao Wang, Jianheng Tang, Yafei Liu, Ruihui Zhao, Wenjie Li, Yefeng Zheng, Xiaodan Liang. (*: Equal Contribution)
IJCAI 2022.
Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step Rewriting [arXiv] [code]
Yi Cheng, Siyao Li, Bang Liu, Ruihui Zhao, Sujian Li, Chenghua Lin, Yefeng Zheng.
ACL 2021.
Unifying Discourse Resources with Dependency Framework [arXiv] [code]
Yi Cheng, Sujian Li, Yueyuan Li.
NLPCC 2021.
Zero-shot Chinese Discourse Dependency Parsing via Cross-lingual Mapping [arXiv]
Yi Cheng, Sujian Li.
DSNNLG workshop@INLG 2019.

Professional Activities

We Media

From December 2020 to July 2022, I served as the editor-in-chief of "夕小瑶科技说" (formerly known as "夕小瑶的卖萌屋"), which is an influential Chinese Wechat public account focused on the field of NLP. During my tenure, I led the team producing over 200 original articles, and the number of subscribers rose from 15k to 100k+. I also wrote 10 original articles myself under the pseudonym of "小轶".

Organizer

I was one of the organizers of the ICLR 2021 Challenge: Machine Learning for Preventing and Combating Pandemics.

Conference Reviewer

I served as the reviewer for ACL (2022), COLING (2022), EMNLP (2021).

Teaching Assistant

I served as the teaching assistant for the course "COMP 6709: Advanced Natural Language Processing" in Spring 2023 at HK PolyU.

Honors

1st Prize for the Competition at 21st China National Conference on Computational Linguistics: Automated Medical Diagnosis Systems in 2021
“Outstanding Graduate of Peking University” in 2020
“Award for Research Excellents” at Peking University in 2018 and 2019
3rd Prize for the “Tencent Cup” Program Design Competition in 2018