---
id: 20261118-T0-13
title: "Read, Route, Patch：让LLM智能体在测试时稳健进化"
title_en: "Read, Route, Patch: Robust Test-Time Evolution for LLM Agents"
url: https://ai.daily.yangsir.net/daily/20261118-T0-13
issue_date: 2026-11-18
publish_date: 2026-11-17T05:00:00.000Z
category: research
source_name: "arXiv cs.CL (NLP)"
source_url: https://arxiv.org/abs/2611.18585
---

# Read, Route, Patch：让LLM智能体在测试时稳健进化

新论文提出 Read, Route, Patch 框架，用于 LLM 智能体的测试时进化，提升长程交互决策能力。现有测试时进化方法存在三个信息瓶颈：记忆检索过度依赖文本相似度、候选筛选不足等。该框架针对这些瓶颈分别处理读取、路由与修补环节。对构建长程任务智能体的开发者来说，这提供了一条在推理阶段持续改进智能体表现、而不必重新训练的路径。

## English Version

**Read, Route, Patch: Robust Test-Time Evolution for LLM Agents**

A new paper proposes the Read, Route, Patch framework for test-time evolution of LLM agents, improving long-horizon interactive decision-making. Existing test-time evolution methods face three information bottlenecks: memory retrieval overemphasizes textual similarity, and candidate screening is insufficient, among others. The framework addresses these bottlenecks across reading, routing, and patching stages. For developers building long-horizon task agents, this offers a path to continuously improve agent performance at inference time without retraining.

---

**来源**：[arXiv cs.CL (NLP)](https://arxiv.org/abs/2611.18585)

**详情页**：https://ai.daily.yangsir.net/daily/20261118-T0-13

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*