Seamus Somerstep
Scholar

Seamus Somerstep

Google Scholar ID: 7UsSuygAAAAJ
University of Michigan
Algorithmic FairnessLLM Alignment
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
37
 
H-index
4
 
i10-index
1
 
Publications
11
 
Co-authors
0
 
Contact
Emailsmrstep@umich.eduGitHubOpen ↗LinkedInOpen ↗
Publications
7 items
A Latent Variable Framework for Scaling Laws in Large Language Models
2025
Cited
0
Limitations of refinement methods for weak to strong generalization
2025
Cited
0
Learning to Choose or Choosing to Learn: Best-of-N vs. Supervised Fine-Tuning for Bit String Generation
2025
Cited
0
CARROT: A Cost Aware Rate Optimal Router
2025
Cited
0
Sloth: scaling laws for LLM skills to predict multi-benchmark performance across families
2024
Cited
0
Microfoundation Inference for Strategic Prediction
arXiv.org · 2024
Cited
0
A transfer learning framework for weak-to-strong generalization
2024
Cited
2
Resume (English only)
Co-authors
0 total
Co-authors: 0 (list not available)