Scholar
Xintong Li
Google Scholar ID: Sw5mq4cAAAAJ
UC San Diego
Citations & Impact
Citations
137
H-index
5
i10-index
2
Publications
14
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- Publications:
- - OCEAN: Offline Chain-of-thought Evaluation and Alignment in Large Language Models (In submission, 2024)
- - CoMMIT: Coordinated Instruction Tuning for Multimodal Large Language Models (In submission, 2024)
- - Open-world Multi-label Text Classification with Extremely Weak Supervision (EMNLP, 2024)
- - Geometry-aware adaptation for pretrained models (NeurIPS, 2023)
- - AutoWS-Bench-101: Benchmarking Automated Weak Supervision with 100 Labels (NeurIPS, 2022)
- News:
- - X-MLClass accepted to EMNLP ‘24 main conference!
- - Started summer internship as an Applied Scientist Intern at Amazon Alexa AI in June 2024
- - Geometry-Aware Adaptation for Pretrained Models accepted to NeurIPS ‘23
- - AutoWS-Bench-101 accepted to NeurIPS ‘22
Research Experience
- Current research includes balancing different modalities in multimodal instruction tuning to improve performance and mitigate catastrophic forgetting.
Education
- Ph.D. candidate at University of California, San Diego (UCSD), advised by Prof. Jingbo Shang; B.S. in Computer Science and Data Science from University of Wisconsin – Madison, where she worked with Prof. Fred Sala and Prof. Jelena Diakonikolas.
Background
- Research Interests: Efficient machine learning, with a particular focus on LLM reasoning, multimodal optimization, and weak supervision. Aims to enhance model reasoning and persona abilities through data-efficient methods and broaden the scope of machine learning methods toward under-studied application fields.
