Ziqian Zhong

PhD student, Computer Science, Carnegie Mellon University

Joined

May 2023

Names

Ziqian Zhong (Preferred)

Emails

****@mit.edu (Confirmed)

****@andrew.cmu.edu (Confirmed)

Personal Links

Homepage

Google Scholar

DBLP

Career & Education History

PhD student

Computer Science, Carnegie Mellon University (cmu.edu)

2025 – Present

Undergrad student

Computer Science, Massachusetts Institute of Technology (mit.edu)

2020 – 2024

Advisors, Relations & Conflicts

PhD Advisor

Aditi Raghunathan

2025 – Present

Coauthor

Jacob Andreas

2022 – 2024

Expertise

AI safety

2025 – Present

Interpretability

2023 – Present

Publications

Early Data Exposure Improves Robustness to Subsequent Fine-Tuning
Lawrence Feng, Gaurav Rohit Ghosal, Jacob Mitchell Springer, Ziqian Zhong, Aditi Raghunathan
- CATS@ICML26 Poster
- Readers: Everyone
Mix Early, Forget Less: Data Mixing During Pretraining Builds Resistance to Forgetting
Lawrence Feng, Gaurav Rohit Ghosal, Jacob Mitchell Springer, Ziqian Zhong, Aditi Raghunathan
- ICLR 2026 Workshop DATA-FM
- Readers: Everyone
Mix Early, Forget Less: Data Mixing During Pretraining Builds Resistance to Forgetting
Lawrence Feng, Gaurav Rohit Ghosal, Jacob Mitchell Springer, Ziqian Zhong, Aditi Raghunathan
- SPOT
- Readers: Everyone
Mix Early, Forget Less: Data Mixing During Pretraining Builds Resistance to Forgetting
Lawrence Feng, Gaurav Rohit Ghosal, Jacob Mitchell Springer, Ziqian Zhong, Aditi Raghunathan
- ICLR 2026 Workshop GRaM Poster
- Readers: Everyone
No One Monitor Fits All: Oversight Strategies for Frontier Agents
Neil Kale, Shashwat Saxena, Ziqian Zhong, Chen Henry Wu, Aditi Raghunathan
- Agentic AI in the Wild: From Hallucinations to Reliable Autonomy Poster
- Readers: Everyone
No One Monitor Fits All: Oversight Strategies for Frontier Agents
Neil Kale, Shashwat Saxena, Ziqian Zhong, Chen Henry Wu, Aditi Raghunathan
- ICLR 2026 AIWILD
- Readers: Everyone
No One Monitor Fits All: Oversight Strategies for Frontier Agents
Neil Kale, Shashwat Saxena, Ziqian Zhong, Chen Henry Wu, Aditi Raghunathan
- ICLR 2026 Trustworthy AI
- Readers: Everyone
Jailbreaking in the Haystack
Rishi Rajesh Shah, Chen Henry Wu, Shashwat Saxena, Ziqian Zhong, Alexander Robey, Aditi Raghunathan
- ICLR 2026 Conference Desk Rejected Submission
- Readers: Everyone
ImpossibleBench: Measuring LLMs' Propensity of Exploiting Test Cases
Ziqian Zhong, Aditi Raghunathan, Nicholas Carlini
- ICLR 2026 Poster
- Readers: Everyone
Watch the Weights: Unsupervised monitoring and control of fine-tuned LLMs
Ziqian Zhong, Aditi Raghunathan
- ICLR 2026 Poster
- Readers: Everyone

View all 19 publications

Ziqian Zhong

Names

Emails

Personal Links

Career & Education History

Advisors, Relations & Conflicts

Expertise

Publications

Co-Authors