Xin Liang

Xin Liang (ๆข้ฆจ)

Incoming PhD Student
Indiana University

Multimodal AI ยท AI Agents ยท Human-Centered AI

Welcome! ๐Ÿ‘‹

I am an incoming PhD student at Indiana University, advised by Prof. Jiangpeng He. I received my B.E. from Tongji University's School of Computer Science (formerly the School of Software Engineering, where my major was Software Engineering) and also spent one year studying at Tongji's College of Architecture and Urban Planning. I have been fortunate to conduct research with Prof. Chenyu You at Stony Brook University, Prof. Chenliang Xu at the University of Rochester, and Prof. Xiaojuan Ma at HKUST. I also interned at StepFun on coding agents and multi-turn dialogue.

My research focuses on multimodal agents and vision-language systems that can reason, plan, and interact with humans in real-world scientific and healthcare workflows.

Latest Updates

News

  1. ๐ŸŽ‰๐ŸŽ† Thrilled to share that SlideGen was accepted to ACM Multimedia 2026! I led this work as first author, what an exciting milestone!
  2. ๐ŸŽ“ Excited to begin my PhD journey at Indiana University, advised by Prof. Jiangpeng He!
  3. ๐Ÿ”ฌ After beginning my internship with Prof. Chenliang Xu's group in 2024, I was delighted to join Prof. Chenyu You's group at Stony Brook University as a research intern!
  4. โœจ Began my research internship with Prof. Chenliang Xu's group at the University of Rochester!
  5. ๐ŸŽ‰ Our work BioMed-VITAL was accepted to NeurIPS 2024! I contributed as the second author.
  6. โ˜€๏ธ Had a wonderful learning experience at a summer school at UC Berkeley!
  7. ๐Ÿค–๐ŸŽ‰ Our robot trajectory labeling work FARPLS was accepted to ACM IUI 2024!
  8. โœจ Excited to join Prof. Xiaojuan Ma's group at HKUST as a research intern!

Selected Work

Research & Projects

ACM MM 2026First Author

SlideGen

Collaborative multimodal agents for automatic scientific slide generation.

Authors: Xin Liang*, Zhilin Zhang*, Xiang Zhang, Haoran Su, Yiwei Xu, Siqi Sun, Chenyu You
Venue: ACM Multimedia (ACM MM), 2026  ยท  * Equal contribution
Project Code Demo
NeurIPS 2024Second Author

BioMed-VITAL

Biomedical visual instruction tuning with clinician preference alignment.

Authors: Hejie Cui*, Lingjun Mao*, Xin Liang, Jieyu Zhang, Hui Ren, Quanzheng Li, Xiang Li, Carl Yang
Venue: NeurIPS 2024, Datasets & Benchmarks Track  ยท  * Equal contribution
View project
ACM IUI 2024Robotics + HCI

FARPLS

A feature-augmented system for labeling robot trajectory preferences.

Authors: Hanfang Lyu, Yuanchen Bai, Xin Liang, Ujaan Das, Chuhan Shi, Leiliang Gong, Yingchi Li, Mingfei Sun, Ming Ge, Xiaojuan Ma
Venue: ACM IUI, 2024
View project
Multi-Agent AIUrban Computing

AI Agent as Urban Planner

Consensus-based multi-agent reinforcement learning for participatory planning.

Authors: Kejiang Qian, Lingjun Mao, Xin Liang, Yimin Ding, Jin Gao, Xinran Wei, Ziyi Guo, Jiajie Li
Venue: arXiv preprint, 2023
View project
Trustworthy InformationWeb3

FactLENS

A decentralized ecosystem for transparent and collaborative news validation.

View project
Virtual RealityLearning

JourneyCam

A VR-assisted photography learning experience with interactive guidance.

View project

Background

Curriculum Vitae

Download PDF

Education

Indiana University Bloomington
PhD Student, advised by Prof. Jiangpeng He
Incoming

Tongji University, Shanghai, China
B.E., School of Computer Science, GPA: 90.4/100, IELTS: 7.0
Formerly the School of Software Engineering; major: Software Engineering.
Also studied for one year at the College of Architecture and Urban Planning.
Sep. 2021 - Jul. 2025

Experience

Research Intern, StepFun

  • Coding agents and multi-turn dialogue systems.

Research Intern, Stony Brook University
Advised by Prof. Chenyu You

  • Sparse adaptation (STAN) with input-dependent routing; consistent gains on LMs/VLMs.
  • SlideGen: multi-agent paper-to-slides pipeline with layout planning & asset matching.

Research Intern, University of Rochester
Advised by Prof. Chenliang Xu

  • Synthetic data & fairness; subgroup-bias analysis across generations.
  • Self-consuming loop to co-train generator/classifier for robust evaluation.

Research Intern, Hong Kong University of Science and Technology (HKUST)
Advised by Prof. Xiaojuan Ma

  • Conducted HCI and human-robot interaction research on robot trajectory preference labeling.
  • Contributed to FARPLS, accepted to ACM IUI 2024.

Research Intern, MIT City Science Lab @ Shanghai

  • Deployed SCRBT token; Solidity contracts on Sepolia with Truffle/Hardhat.
  • Modular smart contracts with ABDK math; end-to-end DApp workflows.

Research Intern, NaMI-Tongji Lab

  • Differentiable NAS via zeroth-order approximation; ICASSP 2023 poster.