Scholar
Yue Wu
Google Scholar ID: LcrSIhgAAAAJ
PhD Student, CMU
Citations & Impact
Citations
3,580
H-index
14
i10-index
16
Publications
19
Co-authors
19
list available
Publications
Resume (English only)
Academic Achievements
- Published a paper titled 'AgentKit: Flow Engineering with Graphs, not Coding,' which proposes an intuitive LLM prompting framework (AgentKit) for multifunctional agents. This work demonstrates SOTA performance of agents designed through AgentKit on WebShop and Crafter. Additionally, co-authored the 'Phi-4 Technical Report.'
Research Experience
- As a PhD student, collaborating with several renowned professors on research involving large language models and reinforcement learning. During undergraduate studies, participated in multiple research projects, including work in the Biorobotics Lab and the MultiComp Lab.
Education
- Currently a PhD student in the Machine Learning Department at Carnegie Mellon University, advised by Dr. Tom Mitchell and Dr. Yuanzhi Li. During undergraduate studies at CMU, conducted research in the Biorobotics Lab under Dr. Howie Choset and in the MultiComp Lab under Dr. Louis-Philippe Morency. Completed an undergraduate thesis on GANs under the supervision of Dr. Zhiting Hu and Dr. Eric Xing. Also worked with Dr. Ruslan Salakhutdinov on sequence generative models in collaboration with Apple Inc.
Background
- Research interests include large language models (LLMs) and reinforcement learning, particularly in embodied reasoning, agent design, and RLHF. Generally interested in solving sequential decision problems with information from unstructured natural language (e.g., instruction manuals or news articles).
Miscellany
- On the market for full-time positions starting September 2024.
