---
format: "aidr-story-markdown/v1"
id: "83fe490d9dc23ecb79c7b0d24df505203c6d5505f0ce224de01568c484831fab"
canonical_url: "https://aidr.today/83fe490d?lang=en"
title: "More Rules, Longer Lists? Benchmarking Self-Evolving Agents on a Hidden-Rule Curriculum"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-02T05:00:00.000Z"
category: "Research"
topics: ["agent"]
source_urls: ["https://arxiv.org/abs/2610.40758"]
summary: "CLR-Bench is a Curriculum of Latent Rules benchmark. Each chain adds one hidden rule per level on top of earlier ones. Twenty chains over 10 domains are released for evaluation of six recent self-evolution methods."
---

# More Rules, Longer Lists? Benchmarking Self\-Evolving Agents on a Hidden\-Rule Curriculum

> [Open the canonical story](<https://aidr.today/83fe490d?lang=en>)

**Published:** 2026-11-02T05:00:00.000Z
**Category:** Research
**Topics:** agent

## Summary

CLR\-Bench is a Curriculum of Latent Rules benchmark\. Each chain adds one hidden rule per level on top of earlier ones\. Twenty chains over 10 domains are released for evaluation of six recent self\-evolution methods\.

## Sources

- [Story source](<https://arxiv.org/abs/2610.40758>)