Doc2Harness: Benchmarking LLMs on Long-Horizon Harness Engineering of Coding Agents
🔗 https://huggingface.co/papers/2610.20344 ↗Doc2Harness is a deterministic two-stage benchmark that evaluates long-horizon harness engineering by having models build and run harnesses across isolated containers, revealing substantial performance gaps among frontier models.
內容為收集當日抓取的摘要原文。