Shuhao (Sullivan) Zhang

M.S. Student

University of California, San Diego

Research Interests

LLM Reasoning
Data-centric AI
Agentic Systems
Reinforcement Learning
Machine Learning

About

I am Shuhao (Sullivan) Zhang, an M.S. student in the Department of Computer Science and Engineering at the University of California, San Diego, advised by Prof. Pengtao Xie.

Previously, I received my B.S. in Data Science from the University of Science and Technology Beijing, where I graduated in the top 7% of my class.

My current research focuses on LLM reasoning, data-centric AI, agentic systems, and reinforcement learning, with a particular interest in building LLM systems that are more reliable, more controllable, and more cost-aware.

News

2026-04 Our paper SCOPE has been accepted to ICML 2026!
2026-04 Preparing our paper SCOUT for submission to EMNLP 2026.
2026-02 We built a project page for SCOPE.
2025-09 Started my M.S. at UC San Diego.
2025-06 Graduated from USTB with a B.S. in Data Science (top 7% of class).

Selected Publications

SCOUT: A Predictor-Guided System for Prompt Injection Detection via Detector Fingerprints

Shuhao Zhang, et al.

Targeting EMNLP 2026 In Preparation

A pre-hoc risk routing framework that predicts attack likelihood and routes untrusted inputs to defense tiers of increasing strength, targeting lower attack success rates with minimal degradation on benign tasks.

Models Under SCOPE: Scalable and Controllable Routing via Pre-hoc Reasoning

Qi Cao^†, Shuhao Zhang^†, Ruizhe Zhou, Ruiyi Zhang, Peijia Qin, Pengtao Xie

ICML 2026 Accepted

SCOPE is a model routing framework that predicts how accurate and how expensive each model will be before running it, enabling controllable cost–accuracy trade-offs and naturally handling new models. Up to 95.1% cost reduction or 25.7% accuracy gain under different priorities.

Project ↗

IDC-CDR: Cross-domain Recommendation based on Intent Disentanglement and Contrast Learning

Jing Xu, Mingxin Gan, Hang Zhang, Shuhao Zhang

Information Processing and Management, 2024

A cross-domain recommendation model that disentangles user intent for stronger transfer, with an emphasis on interpretability of latent intent factors.

Using Machine Learning Models for Short-Term Prediction of Dissolved Oxygen in a Microtidal Estuary

Mina Gachloo, Qianqian Liu, Yang Song, Guozhi Wang, Shuhao Zhang, Nathan Hall

Water 16(14):1998, 2024

Applied ML models for short-term prediction of dissolved oxygen, with relevance to healthcare / biological time-series modeling.