---
format: "aidr-story-markdown/v1"
id: "9b058709d6bcf9aefc3108b7548ad1f3b45cb0f175c3c5f8a44a0147f4debac3"
canonical_url: "https://aidr.today/9b058709?lang=en"
title: "Read, Route, Patch: Robust Test-Time Evolution for LLM Agents"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-17T05:00:00.000Z"
category: "Research"
topics: ["agent"]
source_urls: ["https://arxiv.org/abs/2611.18585"]
summary: "Large Language Model (LLM)-based agents can improve long- horizon interactive decision-making through test-time evolu- tion, yet existing methods face three information bottlenecks: memory retrieval overemphasizes textual similarity, candi- date harnesses are often behaviorally redundant, and whole- harness rewriting may overwrite useful evidence or introduce cross-component inconsistency. We introduce Trident, which improves experience utilization, behavioral exploration, and harness evolution through boundary-aware memory reading, diversity-aware candidate routing, and order-aware insertion- based updating, respectively. We evaluate Trident on five benchmarks spanning text games, web interaction, and cod- ing tasks. Across qwen-32B, mistral-small-3.2-24B, and gpt- oss-20B, Trident achieves the best average performance and consistently outperforms non-learning, memory-based, reflection-based, and automated prompt optimization base- lines. Compared with the strongest baseline, APEX, Trident yields relative average gains of 30.7%, 30.8%, and 38.3% on the three backbones, respectively."
---

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

> [Open the canonical story](<https://aidr.today/9b058709?lang=en>)

**Published:** 2026-11-17T05:00:00.000Z
**Category:** Research
**Topics:** agent

## Summary

Large Language Model \(LLM\)\-based agents can improve long\- horizon interactive decision\-making through test\-time evolu\- tion, yet existing methods face three information bottlenecks: memory retrieval overemphasizes textual similarity, candi\- date harnesses are often behaviorally redundant, and whole\- harness rewriting may overwrite useful evidence or introduce cross\-component inconsistency\. We introduce Trident, which improves experience utilization, behavioral exploration, and harness evolution through boundary\-aware memory reading, diversity\-aware candidate routing, and order\-aware insertion\- based updating, respectively\. We evaluate Trident on five benchmarks spanning text games, web interaction, and cod\- ing tasks\. Across qwen\-32B, mistral\-small\-3\.2\-24B, and gpt\- oss\-20B, Trident achieves the best average performance and consistently outperforms non\-learning, memory\-based, reflection\-based, and automated prompt optimization base\- lines\. Compared with the strongest baseline, APEX, Trident yields relative average gains of 30\.7%, 30\.8%, and 38\.3% on the three backbones, respectively\.

## Sources

- [Story source](<https://arxiv.org/abs/2611.18585>)

