AI & ML interests
None defined yet.
Recent Activity
Papers
Learning Adaptive Reasoning Paths for Efficient Visual Reasoning
From Introspection to Best Practices: Principled Analysis of Demonstrations in Multimodal In-Context Learning
LUKA Lab — Language Understanding and Knowledge Acquisition
Trustworthy Language and Multimodal AI · University of California, Davis
The Language Understanding and Knowledge Acquisition (LUKA) Lab is a research group at the University of California, DavisWe develop robust, generalizable, and accountable machine learning methods for natural language processing, multimodal understanding, and foundation-model agents.
Our long-term goal is to build trustworthy learning systems that help machines understand the world and support scientific discovery.
📚 Research and Open-Source Resources
We share research on reliable and accountable AI, including work on LLM safeguards, model security and provenance, multimodal robustness, knowledge acquisition, and agentic reasoning.