Projects

My projects explore the bidirectional relationship between humans and AI: how AI can enhance human subjective initiative, and how LLMs/agents can better understand human intentions, contexts, and ways of thinking.

Human Subjective Initiative

These projects investigate how AI systems can support reflection, imagination, learning, and creative agency.

Dream Factory

An interactive learning system where users role-play characters from textbook materials with LLM agents to better understand characters, stories, and historical contexts.

AI-Generated New Entities in Collaborative Writing

A theory-driven NLP and linguistics study of how AI-introduced new entities influence meaning-making and the creative process in human-AI collaborative writing.

RewindAI

An AI-clone-based system that helps users revisit past decision points and simulate how paths they did not choose might have unfolded, supporting deeper self-reflection.

Human-like Understanding and Performance

These projects investigate how models and agents can better interpret human intent, context, language, and decision-making signals.

Flexible Smart Glasses

A reconfigurable smart-glasses platform that empowers users to actively configure wearable sensors for context-aware AI interaction, supporting more personalized robot control, physiological awareness, and embodied self-agency.

SNS-Grasp: Semantic-guided Noise Scaling for Grasp Generation

An intent-driven grasp generation framework that uses semantic information to guide robotic grasping.

VagueLlama@SP500

A project exploring how vague language in financial reports reflects management confidence and relates to stock-price trends.