curriculum-learning-design SKILL.md skillA
curriculum-learning-design from anonymousnomad/aide-sovereign-workbench is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add anonymousnomad/aide-sovereign-workbench/curriculum-learning-design@git:20260823.df7cbdePins git:20260823.df7cbde in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- anonymousnomad/aide-sovereign-workbench
- Path
skills/packs/curriculum-learning-design/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260823.df7cbde, committed Aug 23, 2026
- Size
- 6,989 bytes
- Repository stars
- 53
- License
- Apache-2.0
Design and implement curriculum learning for small from-scratch LLM pretraining. Covers difficulty metrics (compression ratio, MTLD, Flesch Reading Ease, code complexity), curriculum strategies (vanilla, pacing-based, interleaved), pacing functions (linear, concave, convex), and CL-as-warmup. Use when designing the pretraining data ordering for the 150M rebuild, choosing difficulty signals, or debugging curriculum-related training instability.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Curriculum Learning Design — 2026 Research-Grounded ## Research Foundations (what actually works for small models) | Source | Proven Principle | Applied As | |--------|------------------|------------| | EACL 2026 (Zhang et al., 2026.eacl-long.271) | **Curriculum learning consistently accelerates convergence 18-45% fewer steps**; compression ratio, MTLD, and Flesch Reading Ease are the most effective difficulty signals; CL-as-warmup yields sustained +3.5% improvement | Our 3-level curriculum (simple/medium/complex) uses compression ratio + token count + FRE; warmup before random sampling | | Readability-driven CL (MDPI 2025) | **Flesch Reading Ease at sentence/group level** with 3 subsets (easy/medium/hard) yields +19.83% on BLiMP; 10 epochs per stage | FRE scoring integrated into Phase 3 filter pipeline; stage boundaries at FRE tertiles | | Curriculum Learning Dynamics (arXiv 2601.21698) | Curricula **reduce gradient noise scale (GNS)** for models 14M-160M; **reverse-order loses gains**; shared latent phases, curricula change phase occupancy | Validate GNS reduction on 150M; never use descending difficulty; monitor phase transitions via HMM | …
Read the whole file at its exact version.
How to install
mdr add anonymousnomad/aide-sovereign-workbench/curriculum-learning-design@sha256:b49921db3a8fb675The line at the top of this page pins the label git:20260823.df7cbde, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_g6dt5ml3xh2f7g7t)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Aug 23, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (6989 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
anonymousnomad/aide-sovereign-workbench · 53 stars · license Apache-2.0 · pushed 2026-10-26 · branch covert-production
API
GET https://markdownregistry.com/api/v1/artifacts/art_g6dt5ml3xh2f7g7t GET https://markdownregistry.com/api/v1/resolve?ref=anonymousnomad/aide-sovereign-workbench/curriculum-learning-design GET https://markdownregistry.com/api/v1/blob/b49921db3a8fb67536ff3f98c6f831dbfef048c7659933b513e796b28cf196b6
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.