[email protected] | twitter | google scholar
🤗 Hey! My name is Abe, which is short for “Abraham” (not the Japanese or Amharic “Abe”). It was given at my birth by my mother who thought it sounded like my Chinese name Bohan Hou (侯博瀚). I am applying for CS (NLP & HCI) PhD program (potentially a joint J.D. or PhD in Social Policy) starting in Fall 2025. I am a fourth-year undergraduate at Johns Hopkins University. I triple major in Computer Science, Sociology, and Mathematics.
🧩 I am broadly interested in exploring the bidirectional interactions between AI and governance, specifically on
(1) How can we build human-centered NLP systems for law and public policy, making governance more efficient and accurate?
(2) How do we interpret and evaluate emergent behaviors of foundation AI models, particularly their reasoning and agentic capabilities?
(3) How do we align AI with pluralistic values while reconciling with safety requirements?
See my detailed explanation (recommended) at
(Recommended) Vision and Long-Term Goals
📝 Selected Publications on Legal NLP:
Abe Bohan Hou, William Jurayj, Nils Holzengerber, Andrew Blair-Stanek, Benjamin Van Durme.
Accepted to NLLP 2024 (EMNLP).
CLERC: A Dataset for Legal Case Retrieval and Retrieval-Augmented Analysis Generation.
Abe Bohan Hou, Orion Weller, Guanghui Qin, Eugene Yang, Dawn Lawrie, Nils Holzenberger, Andrew Blair-Stanek, Benjamin Van Durme
Committed to NAACL 2025 (Meta-review Score: 4). Co-present at NLLP 2024.
L-FRESco: Factual Recall Evaluation Score for Legal Analysis Generation.
Abe Bohan Hou, Zhengping Jiang, Guanghui Qin, Orion Weller, Andrew Blair-Stanek, Benjamin Van Durme
Accepted to ICML Generative Law Workshop 2024.
📝 Selected Publications on AI Safety and Governance:
Stumbling Blocks: Stress Testing the Robustness of Machine-Generated Text Detectors Under Attacks.
Yichen Wang, Shangbin Feng, Abe Bohan Hou, Xiao Pu, Chao Shen, Xiaoming Liu, Yulia Tsvetkov, Tianxing He
Accepted to ACL 2024 Main.
k-SemStamp: A Clustering-Based Semantic Watermark for Detection of Machine-Generated Text.
Abe Bohan Hou, Jingyu Zhang, Yichen Wang, Daniel Khashabi, Tianxing He.
Accepted to ACL 2024 Findings.
SemStamp: A Semantic Watermark With Paraphrastic Robustness For Text Generation.
Abe Bohan Hou,* Jingyu Zhang*, Tianxing He*, Yichen Wang, Yung-Sung Chuang, Hongwei Wang, Lingfeng Shen, Benjamin Van Durme, Daniel Khashabi, Yulia Tsvetkov
Accepted to NAACL 2024 Main.
🤝🏻 At Hopkins, I founded Technology and Policy Society at Johns Hopkins, which focuses on fostering interdisciplinary talents and hosting events related to AI policies. We run the annual Johns Hopkins AI Policy Hackathon in Washington DC with tracks in AI Safety, AI & Public Health, Future of Work, Law, and Sustainability. We are sponsored by OpenAI and Microsoft.
I also serve as the Director of Programming of Johns Hopkins Undergraduate Law Review and have been an editor for three years.