Chinese contentLAGO:把专用语言代理变成可复用选项,SWE-bench 解决 58.6%
新论文 LAGO:把成功子任务存成可复用选项库,不改权重跨任务变强,SWE-bench 上跑赢 mini-SWE-agent 和 Terminus 2。
新论文 LAGO:把成功子任务存成可复用选项库,不改权重跨任务变强,SWE-bench 上跑赢 mini-SWE-agent 和 Terminus 2。
Harness overhead, not model choice, may be the biggest lever on your agent bill—the repeated prompt and schema payloads compound with every step. Teams building coding agents can audit their harness size first before switching models or paying for extra compute. The finding also cautions against treating harness benchmarks as stable model comparisons, since results flip at similar rates across runs.
arXiv 上的新论文 RETRACE,教 AI 智能体从真实 CI 修复历史里学经验,mini-SWE-agent 成功率从 19.6% 涨到 31.9%,做智能体修复的可以看看。