Shihao Weng 翁诗浩

Ph.D. student, Nanjing University

I am a third-year Ph.D. student at Nanjing University, advised by Prof. Yang Feng. I am currently a visiting student at The Chinese University of Hong Kong with Prof. Michael R. Lyu, and before that spent a year at Singapore Management University working with Prof. Xiaofei Xie.

My research is on trustworthy AI: making large language models and agents reliable, safe, and aligned with human values. I work across the model lifecycle, from how models are trained and aligned, to how their behavior is evaluated and understood, to how they are safeguarded once deployed in the real world.

Feel free to reach out at shweng@smail.nju.edu.cn to chat about research or collaboration.

News

Sep 2026Are LLM Safety Judges Policy-Invariant? accepted at NeurIPS 2026 (ED Track) as a Spotlight.
Dec 2025Two papers accepted at ICSE 2026 (Cycle 2).
Jun 2025One paper accepted at ICSE 2026.
Sep 2024One co-authored paper accepted at ISSTA 2024.
Nov 2023One co-authored paper accepted at FSE 2023.
Oct 2023Received the Distinguished Paper Award at Internetware 2023.

Publications † equal contribution * corresponding author

2026

Are LLM Safety Judges Policy-Invariant? A Three-Principle Stress-Test
Shihao Weng, Yang Feng*, Xiaofei Xie

NeurIPS 2026 ED Track · Spotlightpaper & code coming soon
TACO: Trust Assessment of Large Language Models in Coding Assistance Tasks
Shihao Weng, Yang Feng*, Jincheng Li, Yining Yin, Zhenlun Zhang, Lyuxi Liu, Jia Liu

ICSE 2026paper code tool
AtPatch: Debugging Transformers via Hot-Fixing Over-Attention
Shihao Weng, Yang Feng*, Jincheng Li, Yining Yin, Xiaofei Xie, Jia Liu

ICSE 2026paper code
VADA: A Multicultural Benchmark for Value-Aware Data Generation and Alignment Evaluation in LLMs
Zhenlun Zhang, Yang Feng*, Shihao Weng, Yining Yin, Jincheng Li, Jia Liu

ICSE 2026paper code
When Code Becomes an LLM Call: An Empirical Study of Call-Site Contracts in Agentic Systems
Shihao Weng†, Longtian Wang†, Yang Feng*, Mingfei Cheng, Jia Liu, Chao Shen, He Ye, Federica Sarro, Xiaofei Xie

arXiv preprintpaper
AgentAntibody: An Adaptive Immune System for Defending LLM Agents against Prompt Injection
Shihao Weng, Yang Feng*, Xiaofei Xie, Jiongchi Yu

arXiv preprintpaper
ARGUS: Defending LLM Agents Against Context-Aware Prompt Injection
Shihao Weng, Yang Feng*, Jinrui Zhang, Xiaofei Xie, Jiongchi Yu, Jia Liu

arXiv preprintpaper code

2025

Data preparation and quality for code-centric generative software engineering tasks: a systematic literature review
Shihao Weng, Yang Feng*, Yining Yin, Zhenlun Zhang, Baowen Xu

Frontiers of Computer Sciencepaper
Lightweight Probabilistic Coverage Metrics for Efficient Testing of Deep Neural Networks
Yining Yin, Yang Feng*, Shihao Weng, Xinyu Gao, Jia Liu, Zhihong Zhao

Internetware 2025paper code

2024

Datactive: Data Fault Localization for Object Detection Systems
Yining Yin, Yang Feng*, Shihao Weng, Yuan Yao, Jia Liu, Zhihong Zhao

ISSTA 2024paper code
Seeing the invisible: test prioritization for object detection system
Shihao Weng, Yang Feng*, Yining Yin, Yuxuan Dai, Jia Liu, Zhihong Zhao

Empirical Software Engineeringpaper code

2023

Dynamic Data Fault Localization for Deep Neural Networks
Yining Yin, Yang Feng*, Shihao Weng, Zixi Liu, Yuan Yao, Yichi Zhang, Zhihong Zhao, Zhenyu Chen

FSE 2023paper code
Prioritizing Testing Instances to Enhance the Robustness of Object Detection Systems
Shihao Weng, Yang Feng*, Yining Yin, Jia Liu

Internetware 2023 · Distinguished Paper Awardpaper code

Honors

2023–24Outstanding Graduate Student, Nanjing University (×2)
2023National Scholarship for Graduate Students
2023Distinguished Paper Award, Internetware 2023
2021CCF Outstanding Undergraduate Student
2020–21National Scholarship for Undergraduate Students (×2)

Service

ReviewerNeurIPS 2026 Workshop · TAE

Education

2024–nowPh.D. in Software Engineering, Nanjing University, advised by Yang Feng
2022–24M.S. in Software Engineering, Nanjing University
2018–22B.E. in Computer Science, Jiangnan University, advised by Heng-yang Lu