500AI
Back to researchers

Ofir Press

Evaluation + long-context extrapolation

Co-authored SWE-bench and ALiBi: high-leverage evaluation + long-context work.

Highlights

EvaluationSWE-benchALiBiLong context
Focus: Evaluation + long-context extrapolation
Why it matters: Co-authored SWE-bench and ALiBi: high-leverage evaluation + long-context work.
Start here
SWE-bench: Can Language Models Resolve Real-World GitHub Issues?paper →Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation (ALiBi)paper →SWE-bench (GitHub)project →

Research Areas

EvaluationSWE-benchALiBiLong context
Ofir Press - AI Researcher Profile | 500AI