Giulio Starace
Source-listed: Co-founder · Paradigma
- Language models
- Information retrieval
- Evaluation
Selected work
5- PaperBench: Evaluating AI's Ability to Replicate AI ResearchContributor · 2025
- MLE-bench: Evaluating Machine Learning Agents on Machine Learning EngineeringContributor · 2024
- Introducing SWE-bench VerifiedContributor · 2024
- GPT-4o System CardContributor · 2024
- Probing LLMs for Joint Encoding of Linguistic CategoriesContributor · 2023