Multimodal Machine Learning for Group State Sensing - Honda Research Institute USA

Multimodal Machine Learning for Group State Sensing

Your application is being processed

Multimodal Machine Learning for Group State Sensing

Job Number: P24INT-58
​Honda Research Institute USA (HRI-US) is seeking a highly motivated research intern to advance the research on group state sensing. This project focuses on detecting collective states, such as cohesion, morale, engagement, and leadership roles. The states are learned from multimodal data such as speech, gaze, body movement, and physiological signals. The internship is in the intersection of artificial intelligence, human-computer interaction, and social signal processing. The intern will contribute to building machine learning pipelines and models that capture group-level phenomena from individual-level signals, enabling new insights into team performance and human-AI collaboration.
San Jose, CA

 

Key Responsibilities

 
  • Pre-process, synchronize, and extract from heterogeneous data streams
  • Explore embeddings and representation learning methods for multimodal and multi-agent fusion
  • Design and evaluate explainable machine learning architectures
  • Identify emerging behaviors in group state change, such as group-level synchrony and alignment
  • Document research findings clearly and contribute to research publications and presentations      

 

Minimum Qualifications

 
  • Strong M.S. or PhD candidates in Computer Science, Electrical Engineering, Cognitive Science, Robotics, or related field
  • Strong background in machine learning, large language models (LLMs) and data analysis
  • Familiarity with human subject research
  • Strong verbal and written communication skills

 

Bonus Qualifications

  • ​Proven publication record in machine learning, especially in top AI conferences such as Neurips, AAAI, ICLR, ACL, and ICASSP.
  • Demonstrated skills in designing multimodal machine learning (e.g., transformers, fusion methods) and LLM-based architectures (e.g., chain of thought)
  • Knowledge of social signal processing or affective computing
  • Experience with time-series modeling
  • Experience with group-level interaction

 

Years of Work Experience Required  0
Desired Start Date 1/5/2026
Internship Duration 3 Months
Position Keywords

​Multimodal signal processing, behavioral signal, computational sociology

Alternate Way to Apply

Send an e-mail to careers@honda-ri.com with the following:
- Subject line including the job number(s) you are applying for 
- Recent CV 
- A cover letter highlighting relevant background (Optional)

Please, do not contact our office to inquiry about your application status.