🎓 I am on the job market! I am looking for full-time research scientist / applied scientist positions in industry, starting fall 2027. Feel free to reach out at qianqi@ucsb.edu.
I am Qianqi “Jackie” Yan, a Ph.D. candidate in Computer Science @University of California, Santa Barbara, advised by Prof. Xin (Eric) Wang as a member of the UC Santa Barbara AI Group. I have interned as an Applied Scientist at Amazon, Oracle Health AI, and eBay.
I obtained my bachelor’s degrees from the University of Michigan, Ann Arbor (B.S.E. in Computer Science, Summa cum laude) and Shanghai Jiao Tong University (B.S.E. in Electrical and Computer Engineering), where I worked with Prof. Joyce Chai and Prof. Stella Yu.
Research Interests
My research focuses on trustworthy multimodal LLMs and agents. I build benchmarks and attribution methods to measure and mitigate failures of modern multimodal models, including:
- Hallucination & faithfulness: probing evaluations and inconsistency/underspecification reasoning for multimodal LLMs (ProbMed, MMIR, iReason)
- Attribution & auditing: training-free, generation-time evidence attribution across text, image, audio, and video (OmniTrace, clinical summarization auditing)
- Agent safety: deceptive behavior of tool-using LLM agents under pressure (AgentDecept) and agent harness safety
- Generative media forensics & understanding: forgery detection for diffusion-based document edits, and locating reasoning in image/video generators
Please don’t hesitate to email me if you are interested in these areas. I am happy to chat ;)
News
- [2026.11] 🎓 I am on the job market for full-time research/applied scientist roles starting fall 2027!
- [2026.09] Finished my Applied Scientist internship at Amazon, working on generative document forgery detection. Work submitted to CVPR 2027.
- [2026.09] I will be co-organizing the 5th ALVR workshop at ACL 2027.
- [2026.01] PhyWorldBench accepted to ICLR 2026 as an Oral presentation!
- [2025.09] I will be hosting the 4th ALVR workshop at ACL 2026.
- [2025.08] One paper accepted to EMNLP 2025! I will be attending EMNLP 2025 in Suzhou, China. Please feel free to reach out for a chat!
- [2025.05] Two papers accepted to ACL 2025!
- [2025.04] I will join Oracle Health AI as an applied scientist intern in 2025 summer.
- [2024.04] I will join eBay as an applied scientist intern in 2024 summer.
Experience
San Diego, CA · Jun. 2026 – Sep. 2026
Santa Clara, CA · Jun. 2025 – Sep. 2025
San Jose, CA · Jun. 2024 – Sep. 2024
Preprints & Under Review
AgentDecept: Benchmarking Deceptive Behavior in Tool-Using LLM Agents under Pressure
Qianqi Yan, Yehan Zhang, Hang Yin, Ching-Chen Kuo, Shan Jiang, Ming Ouyang, Yijie Liu, Xin Eric Wang
Under reviewOmniTrace: A Unified Framework for Generation-Time Attribution in Omni-Modal LLMs
Qianqi Yan, Yichen Guo, Ching-Chen Kuo, Shan Jiang, Hang Yin, Yang Zhao, Xin Eric Wang
Under review
Do Generators Think in Pixels? Locating Reasoning in Image and Video Generators
Qianqi Yan, Jinru Han, Kaizhi Zheng, Yuhang Gan, Xin Eric Wang
Under reviewDocTwin: A Paired, Factorized Dataset for Generative Document Forgery Detection
Qianqi Yan, Richard Y. Chen, Chi Zhang, Daniel Cociorva, Xin Eric Wang
Under reviewAuditing Agent Harness Safety
Chengzhi Liu, Yichen Guo, Yepeng Liu, Yuzhe Yang, Qianqi Yan, et al.
Under review
Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck
Zihan Huang, Junda Wu, Tong Yu, Qianqi Yan, et al.
Under review
Publications
PhyWorldBench: A Comprehensive Evaluation of Physical Realism in Text-to-Video Models
Jing Gu, Xian Liu, Yu Zeng, Ashwin Nagarajan, Fangrui Zhu, Daniel Hong, Yue Fan, Qianqi Yan, Kaiwen Zhou, Ming-Yu Liu, Xin Eric Wang
ICLR, 2026 Oral
From Attribution to Abstention: Training-Free Attention-Based Auditing for Clinical Summarization
Qianqi Yan, Huy V. Nguyen, Sumana Srivatsa, Hari Bandi, Xin Eric Wang, Krishnaram Kenthapadi
ML4H 2025
Hidden in Plain Sight: Reasoning in Underspecified and Misspecified Scenarios for Multimodal LLMs
Qianqi Yan, Hongquan Li, Shan Jiang, Yang Zhao, Xinze Guan, Ching-Chen Kuo, Xin Eric Wang
EMNLP 2025
Multimodal Inconsistency Reasoning (MMIR): A New Benchmark for Multimodal Reasoning Models
Qianqi Yan, Yue Fan, Hongquan Li, Shan Jiang, Yang Zhao, Xinze Guan, Ching-Chen Kuo, Xin Eric Wang
ACL 2025
Worse than Random? An Embarrassingly Simple Probing Evaluation of Large Multimodal Models in Medical VQA
Qianqi Yan, Xuehai He, Xiang Yue, Xin Eric Wang
ACL 2025
Muffin or Chihuahua? Challenging Large Vision-Language Models with Multipanel VQA
Yue Fan, Jing Gu, Kaiwen Zhou, Qianqi Yan, Shan Jiang, Ching-Chen Kuo, Xinze Guan, Xin Eric Wang
ACL 2024
Med-HVL: Automatic Medical Domain Hallucination Evaluation for Large Vision-Language Models
Qianqi Yan, Xuehai He, Xin Eric Wang
AAAI 2024 Spring Symposium on Clinical Foundation Models
Service
Organizer
- Workshop: 5th ALVR @ACL 2027, 4th ALVR @ACL 2026, CLVL @ICCV 2025
Editor
- Proceedings of the 4th Workshop on Advances in Language and Vision Research (ALVR), ACL 2026
Reviewer
- Main Conference: NeurIPS 2026, ICLR 2024-2026
- Workshop: ALVR @ACL 2024, CLVL @ICCV 2023
Education
- Ph.D. in Computer Science, University of California, Santa Barbara, 2025 - Jun. 2027 (expected)
- Ph.D. in Computer Science, University of California, Santa Cruz, 2023 - 2025 (transferred with advisor)
- B.S.E. in Computer Science, Summa cum laude, University of Michigan (Ann Arbor), 2021 - 2023
- B.S.E. in Electrical and Computer Engineering, Shanghai Jiao Tong University, 2019 - 2023
- Visiting Student, Technische Universität Berlin, 2020
Teaching
- CS165A: Artificial Intelligence, University of California, Santa Barbara
- CSE101: Introduction to Data Structures and Algorithms, University of California, Santa Cruz
- EECS 448: Human-Centered Machine Learning, University of Michigan, Ann Arbor
- VE 215: Intro to Circuits, Shanghai Jiao Tong University
- VG 101: Intro to Computers and Programming, Shanghai Jiao Tong University