跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.10982 · v1

Think or Remember: The Exchange Rate of Agentic Inference

首次提交 2026-11-08 05:45
版本更新 2026-11-08 05:45
雷达观测 2026-11-10 12:16
arXiv ↗PDF ↗

摘要

Agents built on reasoning models resend their growing transcript at every call. Two transacting parties, not the model, decide what that costs: the client keeps or edits messages; the server decides whether earlier reasoning returns in the prompt and what its cache reuses. We define a cost model whose counts match an instrumented engine exactly; the architecture charges cc per token processed and ββ per pair attended. Their ratio P∗=c/βP^*=c/β is an exchange rate between thinking and remembering: per token, processing dominates below this context length, a thinking regime, and attention above it, a memory regime. A policy change has a mechanical cost, incurred even if the agent writes the same tokens, and a behavioral one: it writes different tokens, changing cost and outcome. Serving alone can flip which policy is cheaper: on recorded Qwen3-8B ScienceWorld runs, dropping earlier reasoning costs 1.74 times keeping it when the server reuses generated states and 0.34 times when it reuses nothing. In live reruns, dropping it cuts correct HotpotQA answers by 9-23 percentage points on four models, at lower point-estimate cost. Comparing context policies means comparing systems: architecture and serving set the cost of the history kept and the work it induces.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-08 05:45 · 最新观测版本

Think or Remember: The Exchange Rate of Agentic Inference

Agents built on reasoning models resend their growing transcript at every call. Two transacting parties, not the model, decide what that costs: the client keeps or edits messages; the server decides whether earlier reasoning returns in the prompt and what its cache reuses. We define a cost model whose counts match an instrumented engine exactly; the architecture charges cc per token processed and ββ per pair attended. Their ratio P∗=c/βP^*=c/β is an exchange rate between thinking and remembering: per token, processing dominates below this context length, a thinking regime, and attention above it, a memory regime. A policy change has a mechanical cost, incurred even if the agent writes the same tokens, and a behavioral one: it writes different tokens, changing cost and outcome. Serving alone can flip which policy is cheaper: on recorded Qwen3-8B ScienceWorld runs, dropping earlier reasoning costs 1.74 times keeping it when the server reuses generated states and 0.34 times when it reuses nothing. In live reruns, dropping it cuts correct HotpotQA answers by 9-23 percentage points on four models, at lower point-estimate cost. Comparing context policies means comparing systems: architecture and serving set the cost of the history kept and the work it induces.

查看此版本解读 →查看此版本原文 ↗