Xingda Lyu (David)

Hi! I'm an undergraduate researcher at the University of Washington in the Computational Minds and Machines Lab with Prof. Max Kleiman-Weiner. I was a summer research intern in the Data Mining Group at UIUC, advised by Prof. Jiawei Han. Previously, I worked with Prof. Chirag Shah at the UW InfoSeeking Lab.

My research focuses on proactive and adaptive AI agents. I'm interested in how agents can recognize gaps in users' understanding, decide when and how to help, and improve through feedback and experience. My work spans human-AI collaboration, scientific discovery, and multimodal generation.

Click here to view my CV.

News

2026.09: Attended the Human-AI Complementarity Workshop (CMU AI-SDM) in Pittsburgh, PA.
2026.07: Attended ICML 2026 in Seoul.
2026.06: Attended CVPR 2026 in Denver, Colorado.
2026.06: Started a summer research internship with Prof. Jiawei Han at UIUC.
Show moreShow less
2026.05: Our paper Knowing Isn't Understanding was accepted to ICML 2026.
2026.04: Our paper PROS was accepted to the CVPR HiGen Workshop.

Research Experience

  1. Aug 2026 – Present

    Computational Minds and Machines Lab logo

    Computational Minds and Machines Lab

    University of Washington

    Research Assistant

    Advisor: Max Kleiman-Weiner

    PhD mentor: Kirandeep Kaur

  2. Jun – Aug 2026

    UIUC Data Mining Group logo

    Data Mining Group

    University of Illinois Urbana-Champaign

    Summer Research Intern

    Advisor: Jiawei Han

    PhD mentor: Joey Chan

  3. Nov 2025 – Aug 2026

    InfoSeeking Lab logo

    InfoSeeking Lab

    University of Washington

    Research Assistant

    Advisor: Chirag Shah

    PhD mentor: Kirandeep Kaur

Selected Publications

Preprint Before-and-after poster refinement examples showing accepted structural, scientific, and spatial repairs

Beyond Instruction-Driven Editing: Source-Grounded Problem Discovery with User-Governed Repair for Scientific Posters

Xingda Lyu, Honglin Lu, Xinye Luo, and Shiqi Yang.

CVPR HiGen Workshop, 2026.

ICML ’26 Layered epistemic landscape showing knowns, unknowns, and a proactive agent surfacing knowledge gaps

Knowing Isn't Understanding: Re-grounding Generative Proactivity with Epistemic and Behavioral Insight

Kirandeep Kaur, Xingda Lyu, and Chirag Shah.

International Conference on Machine Learning (ICML), 2026.

NeurIPS ’26 Agents' Last Exam overview showing professional task domains and real-world workflows

Agents' Last Exam

Yiyou Sun, …, Xingda Lyu, …, Dawn Song.

Conference on Neural Information Processing Systems (NeurIPS), 2026.

CONF-CDS ’25 P-RAG paper first-page preview showing the title, authors, affiliations, and opening of the abstract

P-RAG: Prompt-Enhanced Parametric RAG with LoRA and Selective CoT for Biomedical and Multi-Hop QA

Xingda Lyu, Gongfu Lyu, Zitai Yan, and Yuxin Jiang.

International Conference on Computing and Data Science (CONF-CDS), 2025.

Posters & Talks

CHIIR ’26 Preview for the CHIIR proactive agent presentation

Decision-Making Under Unknown Unknowns in Proactive Agentic Systems

Xingda Lyu.

ACM CHIIR Workshop on Human-Centered Proactive & Personalized Agents, 2026.

ARD ’26 ProPer framework showing user queries, explicit and implicit dimensions, reranking, response generation, and feedback from successful interactions

ProPer: Proactivity-Driven Personalized Agents for Knowledge Gap Navigation

Kirandeep Kaur, Vinayak Gupta, Xingda Lyu, Aditya Gupta, and Chirag Shah.

Amazon Research Day, AI Co-Scientist, 2026.

Industry Experience

Amazon

Data Science Intern

2025.06 - 2025.08 Seattle, WA

Baynovation

Data Science Intern

2024.07 - 2024.09 San Jose, CA

Lab Photos

Computational Minds and Machines Lab, 2026
1 / 2

Outside research, I play competitive badminton.

Lab Photos

Computational Minds and Machines Lab group photo outdoors, 2026
Computational Minds and Machines Lab, 2026
1 / 2