Benjamin Cohen-Wang
Scholar

Benjamin Cohen-Wang

Google Scholar ID: QwJR7jEAAAAJ
PhD Student, MIT
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
110
 
H-index
5
 
i10-index
5
 
Publications
8
 
Co-authors
0
 
Contact
Emailbencw@mit.eduTwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
2 items
Learning to Attribute with Attention
2025
Cited
0
SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models
2025
Cited
0
Resume
Academic Achievements
  • Selected papers:
  • - ContextCite: Attributing Model Generation to Context (NeurIPS 2024)
  • - Ask Your Distribution Shift if Pre-Training is Right for You (TMLR 2025)
  • - Learning to Attribute with Attention (Preprint)
  • Blog posts and software projects:
  • - Using ContextCite for LLM reliability
  • - ContextCite: Attributing Model Generation to Context
  • - How Can We Harness Pre-Training to Develop Robust Models?
  • - Ask Your Distribution Shift if Pre-Training is Right for You
  • - Attribution with Attention (AT2): Efficiently attribute generated content to in-context information
  • - Software tools include GitHub repository, Hugging Face demo, and PyPI package.
Research Experience
  • Worked at Robust Intelligence for two years before starting his PhD at MIT.
Education
  • BS in Computer Science from Stanford University; PhD in EECS at MIT, advisor Aleksander Mądry.
Background
  • Research Interests: Developing AI systems that can be safely deployed. Background: Currently a PhD student in EECS at MIT, advised by Aleksander Mądry.
Miscellany
  • Personal interests: Running, climbing, skiing, tennis, and volleyball.