NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents Paper • 2512.12730 • Published 12 days ago • 43
FinSearchComp: Towards a Realistic, Expert-Level Evaluation of Financial Search and Reasoning Paper • 2509.13160 • Published Sep 16 • 29
Inverse IFEval: Can LLMs Unlearn Stubborn Training Conventions to Follow Real Instructions? Paper • 2509.04292 • Published Sep 4 • 57