Scholar
Xiao Yu
Google Scholar ID: QblBy88AAAAJ
Computer Science PhD Student, Columbia University
Citations & Impact
Citations
192
H-index
8
i10-index
6
Publications
17
Co-authors
6
list available
Publications
Resume (English only)
Academic Achievements
- arXiv: Dyna-Mind: Learning to Simulate from Experience for Better AI Agents
- arXiv: Dyna-Think: Synergizing Reasoning, Acting, and World Model Simulation in AI Agents
- NeurIPS 2025 (Workshop): AI Agents for Web Testing: A Case Study in the Wild
- ICLR 2025: ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning
- ACL 2025: ConFit v2: Improving Resume-Job Matching using Hypothetical Resume Embedding and Runner-Up Hard-Negative Mining
Research Experience
- Recent work includes papers published on arXiv like 'Dyna-Mind: Learning to Simulate from Experience for Better AI Agents' and others.
Education
- Third-year Ph.D. student in Computer Science at Columbia University, advised by Zhou Yu; Undergraduate degree from Columbia University, majoring in Computer Science and minoring in Applied Physics.
Background
- Research interests: improving the environment understanding and planning capabilities of AI agents, especially for browser/computer/phone-use. Focus on scalable reinforcement learning algorithms, world model training methods such as Dyna, and planning algorithms such as MCTS. Professional field: Computer Science.
