[CfP] ICML 2025 Workshop on Models of Human Feedback for AI Alignment

[Thomas Kleine Buening]

We are pleased to announce the 2nd Models of Human Feedback for AI Alignment (MoFA) Workshop at ICML 2025 taking place July 18, 2025 in Vancouver, Canada. 

The workshop will discuss key questions for AI alignment and learning from human feedback including how to model human feedback, how to learn from diverse human feedback, how to learn from synthetic feedback, and how to ensure alignment despite misspecified feedback models. 

Call for Papers:  https://sites.google.com/view/mhf-icml2025/call-for-papers 
Submission Portal:  https://openreview.net/group?id=ICML.cc/2025/Workshop/MoFA

Key dates:

• Submission Deadline: May 25 AoE, 2025
• Acceptance Notification: June 9, 2025
• Workshop: July 18, 2025 

We invite submissions related to the theme of the workshop.Topics and relevant research areas include but are not limited to:

• Learning from Demonstrations 
• Reinforcement Learning from Human Feedback
• Preference Learning 
• Human-AI Alignment 
• Human-Robot Collaboration 
• Learning from Synthetic / AI-generated Feedback
• Computational Social Choice 
• Behavioral Economics (e.g., bounded rationality)
• Cognitive Science

Best,
The Organizers 

(Belen, Micah, Maria, Thomas, Andreas, Anca)