Evolver：让 Claude Code 和 Codex 复用历史轨迹，成功率提升至 87.50%
一个叫 Evolver 的框架，把 Claude Code 和 Codex 跑过的经验存下来复用，不用改模型，成功率涨到 87.50%，token 还省了 24%。
Evolver 是一个把智能体探索过程转化为可复用经验的框架，经验以结构化 "Gene" 资产存储，任务开始时检索、事后根据反馈修订。在不修改模型参数的前提下，Claude Code 和 Codex 复用先前运行学到的知识后，在 Terminal-Bench 2.1、SWE-bench Multilingual、SWE-rebench 和 GAIA 的 320 组配对运行中，任务成功率从 83.13% 升至 87.50%。记录的 token 用量下降 24.24%；在 265 个有无可不用的对照任务上，token 仍下降 17.57%。
Impressive way to reuse agent trajectories to evolve agent harnesses. Given how good LLM agents are getting at context awareness, this is worth a look. This idea isn't new, but it's becoming an important pattern in harness engineering: enabling the agent harness to reuse accumulated knowledge without modifying model parameters. They introduce Evolver, a framework for transforming transient agent exploration into persistent, reusable procedural experience. They find that Claude Code and Codex solve more tasks and use fewer tokens when they can reuse what earlier runs learned, with no change to the models. The framework stores that procedural knowledge as structured "Gene" assets. Evolver collects them during exploration, retrieves the relevant ones at the start of a task, and revises them based on feedback. On a selected diagnostic set of 320 paired runs across Terminal-Bench 2.1, SWE-bench Multilingual, SWE-rebench, and GAIA, task success rises from 83.13% to 87.50%. Recorded token use drops by 24.24%. On the 265 tasks solved both with and without Evolver, tokens still drop by 17.57%. The agent spends fewer tokens to reach the same result. arxiv.org/abs/2611.07595 Chat with Paper: academy.dair.ai/papers/evolver… 💬 16 🔄 0 ❤️ 9 👀 1439 📊 15 ⚡
