Ofir Press
Source-listed: Research Scientist · Meta
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- ProgramBench: Can Language Models Rebuild Programs From Scratch?Saved credit: Contributor · 2026Explore people connected to this work →
- SWE-bench: Can Language Models Resolve Real-World GitHub Issues?Saved credit: Contributor · 2024Explore people connected to this work →
- SWE-agent: Agent-Computer Interfaces Enable Automated Software EngineeringSaved credit: Contributor · 2024Explore people connected to this work →
- Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation (ALiBi)Saved credit: Creator · 2022Explore people connected to this work →
- Measuring and Narrowing the Compositionality Gap in Language ModelsSaved credit: Contributor · 2023Explore people connected to this work →