Alexander Wettig
Princeton NLP PhD; co-author on SWE-bench, ShortLong-Doc, and the QuRating / data-selection line of pretraining-data work.
- Role
- researcher
- Currently at
- Princeton NLP Group
- twitter.com/_awettig
- GitHub
- github.com/awettig
- Scholar
- scholar.google.com/citations
- Papers
- 12
Cite
Notes
Only stored in your browser.
Authored papers
12SWE-smith: Scaling Data for Software Engineering Agents
arXiv 2025
Olmo 3
arXiv 2025
Metadata Conditioning Accelerates Language Model Pre-training
arXiv 2025
SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
ICLR
OLMoE: Open Mixture-of-Experts Language Models
arXiv 2024
How to Train Long-Context Language Models (Effectively)
arXiv 2024
QuRating: Selecting High-Quality Data for Training Language Models
arXiv 2024
Language Models as Science Tutors
arXiv 2024
Adapting Language Models to Compress Contexts
arXiv 2023
Learning Transformer Programs
learning-transformer-programs
A Kernel-Based View of Language Model Fine-Tuning
arXiv 2022
Should You Mask 15% in Masked Language Modeling?
arXiv 2022
Affiliations
Frequent co-authors
10from 12 papers
Danqi Chen
professor
Tianyu Gao
Akshita Bhagia
Alexis Chevalier
Ali Farhadi
CEO
Binyuan Hui
Dirk Groeneveld
Dustin Schwenk
Hannaneh Hajishirzi
professor
Jacob Morrison
research-engineer