aifollow.news
返回 Rohan Paul
Rohan Paul· @rohanpaul_ai · X· · 原发布时间 AI 评分54

TokenEconomizer 研究用外部控制器降低编程智能体的令牌成本

AI 导读

一篇来自 UC Berkeley 和 Microsoft 的论文提出,让外部控制器从编程智能体的历史运行中学习,按需注入事实、移除未使用的工具,并提前终止失败循环。在 Claude Code、Codex 和 GitHub Copilot CLI 的 277 项 SWE-Bench 任务测试中,报告称平均每项任务成本降低 18.4% 至 30.9%。

原帖引用 · 原文

@rohanpaul_ai

– https://t.co/rBBRuMoum2

Title: "TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents"

回复Rohan Paul@rohanpaul_ai
New UC Berkeley and Microsoft paper cuts coding-agent token costs by 18.4% to 30.9% using an external controller that learns from past runs. by injecting learned facts, pruning unused tools, and stopping failed loops early. Coding agents are billed by the token, and each turn resends the growing history, so long projects get expensive. Token efficiency is usually measured on isolated tasks, missing the cost that builds up across a sequence of tasks in the same codebase. TokenEconomizer sits outside the agent as a black box, changing only prompts and tool settings. From past runs it learns which facts to inject, which unused tools to drop, and when to stop a failing loop. Across Claude Code, Codex, and GitHub Copilot CLI on 277 SWE-Bench tasks, it cut mean per-task cost 18.4% to 30.9%. It improved Pass@1 in 4 of 5 agent-model settings, and savings held in every repository. Measure agent cost across a whole project, not per task, and use an external controller that adapts to each repo, not fixed plugins.
在 X 查看这条帖子 ↗

来源:Rohan Paul · x.com

论文
发现内容有误?提交纠错