About
Hello and welcome! I am a master’s student in the Data Mining Lab at KAIST, where I focus on understanding the collective behavior of state-of-the-art generative models. My research is centered around key areas like social network mining and large language models (LLMs).
After earning my bachelor’s degree in Electrical Engineering and Computer Science from GIST, I interned at Yonsei University’s Vision and Learning Lab. There, I collaborated with professor Jonghyun Choi and researcher Byungwook Kim, contributing to an ICCV conference paper on autonomous robot planning for interaction and navigation using natural language.
My primary research question revolves around the behavioral implications of advanced generative AI models. Through this blog, I look forward to sharing insights from my academic journey and explorations in this exciting field.
Thank you for visiting, and I hope you find my work and thoughts engaging!
Publication
Conference
ICCV Context-Aware Planning and Environment-Aware Memory for Instruction Following Embodied Agents
- Won 1st prize in ALFRED challenge
- Presented at: ICCV 2023
- Authors: Byeonghwi Kim, Jinyeon Kim, Yuyeong Kim, Cheolhong Min, Jonghyun Choi
- Summary: Our paper introduces the CPEM (Context-aware Planner and Environment-aware Memory) embodied agent, revolutionizing AI agents’ ability to perform household tasks. CPEM’s unique approach to integrating contextual data and maintaining updated spatial object arrangements results in significantly enhanced visual navigation and object interaction capabilities. This method has shown exceptional improvements in task success across various challenging benchmarks, particularly in new and unfamiliar environments.
Workshop
CVPRW Generalist Language Grounding Agents Challenge
- Won 1st prize
- Presented at: CVPRW 2023