Sara Hooker
VP of Research at Cohere and head of Cohere Labs; multilingual NLP and eval-critique researcher; co-author of the "Leaderboard Illusion" paper.
- Role
- researcher / VP Research
- Currently at
- Cohere
- twitter.com/sarahookr
- GitHub
- github.com/sarahooker
- Scholar
- scholar.google.com/citations
- Papers
- 11
Cite
Notes
Only stored in your browser.
Authored papers
11The Leaderboard Illusion
preprint
Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluation
arXiv 2025
Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model
ACL
M-RewardBench: Evaluating Reward Models in Multilingual Settings
arXiv 2024
From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models
arXiv 2024
LLM See, LLM Do: Guiding Data Generation to Target Non-Differentiable Objectives
arXiv 2024
On the Challenges of Using Black-Box APIs for Toxicity Evaluation in Research
arXiv 2023
Pushing Mixture of Experts to the Limit: Extremely Parameter Efficient MoE for Instruction Tuning
arXiv 2023
The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI
arXiv 2023
The Goldilocks of Pragmatic Understanding: Fine-Tuning Strategy Matters for Implicature Resolution by LLMs
the-goldilocks-of-pragmatic-understanding
What Do Compressed Deep Neural Networks Forget?
arXiv 2019
Eval contributions
1Tool contributions
1Affiliations
Previously
Frequent co-authors
10from 11 papers
Beyza Ermiş
Marzieh Fadaee
researcher
Ahmet Üstün
researcher
Shayne Longpre
researcher
Shivalika Singh
engineer
Drishti Sharma
Julia Kreutzer
researcher
Luísa Shimabucoro
Luiza Pozzobon
Niklas Muennighoff
grad-student