1 paper
Atharva Sehgal, James Hou, Akanksha Sarkar +4
Large language model (LLM) coding agents increasingly operate at the repository level, motivating benchmarks that evaluate their ability to optimize entire codebases under realisti…