Scientific Frontiers

Physical AI and Robotic Foundation Models

This research pillar builds integrated foundation models for robot autonomy, unifying perception, decision-making, and physical control. Grounding vision, language, and sensory data in embodied intelligence, they let robots reason in context and act safely in unpredictable settings — across smart manufacturing, logistics, and assistive robotics.

Mosaic3D — open-vocabulary 3D scene segmentation with dense language labels — From “Mosaic3D: Foundation Dataset and Model for Open-Vocabulary 3D Segmentation” · CVPR 2025

Publications List

Foundation Model for Robotics

# Foundation Model for Robotics

MemBench: Memorized Image Trigger Prompt Dataset for Diffusion Models

Chunsan Hong, Tae-Hyun Oh, Minhyuk Sung · TMLR · 2025

# Foundation Model for Robotics

VLM’s Eye Examination: Instruct and Inspect Visual Competency of Vision Language Models

Nam Hyeon-Woo, Moon Ye-Bin, Wonseok Choi, Lee Hyun, Tae-Hyun Oh · TMLR · 2025

# Foundation Model for Robotics

Physical AI and Robotic Foundation Models

Publications List

MemBench: Memorized Image Trigger Prompt Dataset for Diffusion Models

VLM’s Eye Examination: Instruct and Inspect Visual Competency of Vision Language Models

Enhancing Cost Efficiency in Active Learning with Candidate Set Query

A Unified Framework for Motion Reasoning and Generation in Human Interaction

Quantized Factor Identifiable Causal Effect Variational Autoencoder

Frequency-Conditioned Diffusion Models for Time Series Generation