About me

Hi, I’m Kaylee. I am a third-year Ph.D. student in Computer Science at University of California, San Diego. I am fortunate to be advised by Prof. Jingbo Shang. Previously, I received my B.S. degree in Computer Science and Data Science at University of Wisconsin – Madison, where I worked with Prof. Fred Sala and Prof. Jelena Diakonikolas.

My research focuses on efficient machine learning, with a particular interest in reasoning and multimodality. I aim to enhance multimodal reasoning ability by encouraging models to leverage non-textual modalities more effectively. I have also developed methods to mitigate hallucination and reduce catastrophic forgetting during multimodal instruction tuning. Beyond multimodal reasoning, my broader work in reinforcement learning based post-training focuses on improving reasoning efficiency and stability in both text-only and multimodal models. Over the past summer, I have been developing methods for long-to-short reasoning.

Please get in touch with me via email if you would like to discuss research or potential collaborations!

News

• Sep 2025: Four papers are accepted to EMNLP ‘25 conference!
• Jun 2025: I have started my summer internship as an Applied Scientist Intern at Amazon Rufus Team!

Selected Publications

• Importance Sampling for Multi-Negative Multimodal Direct Preference Optimization
  Xintong Li*, Chuhan Wang*, Junda Wu, Rohan Surana, Tong Yu, Julian McAuley, Jingbo Shang.
  In submission, 2025.
  [Paper] [Code]

• Toward Multi-Session Personalized Conversation: A Large-Scale Dataset and Hierarchical Tree Framework for Implicit Reasoning
  Xintong Li, Jalend Bantupalli, Ria Dharmani, Yuwei Zhang, Jingbo Shang.
  EMNLP 2025 (Oral).
  [Paper] [Code] [Dataset]

• CoMMIT: Coordinated Multimodal Instruction Tuning
  Xintong Li*, Junda Wu*, Tong Yu, Rui Wang, Yu Wang, Xiang Chen, Jiuxiang Gu, Lina Yao, Julian McAuley, Jingbo Shang.
  EMNLP 2025 [Paper] [Code]

• Mitigating Visual Knowledge Forgetting in MLLM Instruction-tuning via Modality-decoupled Gradient Descent
  Junda Wu, Yuxin Xiong, Xintong Li, Yu Xia, Ruoyu Wang, Yu Wang, Tong Yu, Sungchul Kim, Ryan A Rossi, Lina Yao, Jingbo Shang, Julian McAuley.
  EMNLP Findings 2025 [Paper] [Code]

• Explainable Chain-of-Thought Reasoning: An Empirical Analysis on State-Aware Reasoning Dynamics
  Sheldon Yu, Yuxin Xiong, Junda Wu, Xintong Li, Tong Yu, Xiang Chen, Ritwik Sinha, Jingbo Shang, Julian McAuley.
  EMNLP Findings 2025 [Paper] [Code]

• OCEAN: Offline Chain-of-thought Evaluation and Alignment in Large Language Models
  Junda Wu*, Xintong Li*, Ruoyu Wang, Yu Xia, Yuxin Xiong, Jianing Wang, Tong Yu, Xiang Chen, Branislav Kveton, Lina Yao, Jingbo Shang, Julian McAuley.
  ICLR 2025.
  [Paper] [Code]

• Open-world Multi-label Text Classification with Extremely Weak Supervision
  Xintong Li, Jinya Jiang, Ria Dharmani, Jayanth Srinivasa, Gaowen Liu, Jingbo Shang.
  EMNLP 2024.
  [Paper] [Code]

• Geometry-aware adaptation for pretrained models
  Nicholas Roberts, Xintong Li, Dyah Adila, Sonia Cromp, Tzu-Heng Huang, Jitian Zhao, Frederic Sala.
  NeurIPS 2023.
  [Paper]

• AutoWS-Bench-101: Benchmarking Automated Weak Supervision with 100 Labels
  Nicholas Roberts*, Xintong Li*, Tzu-Heng Huang, Dyah Adila, Spencer Schoenberg, Cheng-Yu Liu, Lauren Pick, Haotian Ma, Aws Albarghouthi, Frederic Sala.
  NeurIPS 2022.
  [Paper] [Code]