Research Projects

  • R2 - Quantitative Biology
  • M2 - Vision and Motion

Raman Spectroscopy Had No Benchmark. So We Made One.

  • R1 - Robotics
  • M2 - Vision and Motion

3D Hand Shape Estimation for Video-to-Texture Mapping

  • R2 - Quantitative Biology

AI-based interactive segmentation approaches for annotating biomedical image data

  • R1 - Robotics
  • M2 - Vision and Motion

Embodied AI: Vision-Language-Action Models

  • R1 - Robotics
  • M2 - Vision and Motion

Finger Contact Forces from Thermal Signatures

  • R1 - Robotics

Lagrangian Neural Networks for Computed Torque Control

  • M3 - Evaluation, Fairness, Robustness

LLMStructBench – Structured JSON data extraction benchmark and evaluation

  • R2 - Quantitative Biology
  • M1 - Continual Few-Shot Learning
  • M2 - Vision and Motion

Morphological Fingerprints of Antibiotic Resistance

  • M4 - Adapting Text Embeddings

Multimodal idiom detection

  • R3 - Predictive Medicine
  • M4 - Adapting Text Embeddings

Predicting Clinical Outcomes

  • R1 - Robotics

Real-Time Learning Control for Quadruped Robot Velocity Tracking

  • M3 - Evaluation, Fairness, Robustness

Resilient Time Series Models for Urban Wastewater Management

  • M4 - Adapting Text Embeddings

Self-Improving Systems

  • R1 - Robotics

SIRo Soft Hand

  • R2 - Quantitative Biology
  • M2 - Vision and Motion

Synthetic Data Enables Human-Grade Microtubule Segmentation with Foundation Models

  • R2 - Quantitative Biology

TabMetaEns: Meta-learning ensemble portfolios for tabular domains with structured statistical fingerprints

  • R1 - Robotics

What makes a good robot?