Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Mbzuai oryx Awesome LLM Post training Taxonomy Section Structure

From Leeroopedia


Knowledge Sources
Domains Curation, Information_Architecture
Last Updated 2026-02-08 07:30 GMT

Overview

Concrete pattern for defining the hierarchical section structure and table of contents in the awesome-list README.

Description

The taxonomy in README.md is implemented as markdown section headers (H2/H3) organized into a table of contents at lines 33-48. The categories are derived from the companion survey paper (arXiv:2502.21321) and its taxonomy figure (Images/teasor.jpg). The structure uses a two-level hierarchy: major sections (Papers, LLMs in RL, Reward Learning, Policy Optimization, etc.) with subsections (Survey, Theory, Explainability, etc.).

This is a Pattern Doc as it documents a manual editorial process, not a programmatic API.

Usage

Define this structure before beginning paper categorization. The table of contents provides anchor links to each section for in-page navigation. Section names should match the survey paper's taxonomy to maintain consistency.

Code Reference

Source Location

  • Repository: Awesome-LLM-Post-training
  • File: README.md
  • Lines: 33-48 (table of contents), full section headers throughout the file

Pattern Specification

## Contents Table Format (Markdown)

| Section | Subsection |
| ------- | ----------- |
| [Papers](#papers) | [Survey](#survey), [Theory](#theory), [Explainability](#explainability) |
| [LLMs in RL](#LLMs-in-RL) | LLM-Augmented Reinforcement Learning |
| [Reward Learning](#reward-learning) | [Human Feedback], [Preference-Based RL], [Intrinsic Motivation] |
| [Policy Optimization](#policy-optimization) | [Offline RL], [Imitation Learning], [Hierarchical RL] |
...

Section Header Format

# Top-level: Papers
## Second-level: Survey, Theory, Explainability
## Second-level: LLMs in RL
## Second-level: Reward Learning
### Third-level: Human Feedback, Preference-Based RL, Intrinsic Motivation

I/O Contract

Inputs

Name Type Required Description
Survey paper taxonomy Reference Yes The taxonomy figure and structure from arXiv:2502.21321
Domain knowledge Editorial Yes Understanding of LLM post-training research landscape

Outputs

Name Type Description
Table of contents Markdown Linked table at top of README mapping sections to anchors
Section headers Markdown H2/H3 headers throughout README defining category boundaries

Usage Examples

Taxonomy Categories Used

Major Categories (from survey paper):
1. Papers (Survey, Theory, Explainability)
2. LLMs in RL
3. Reward Learning (Human Feedback, Preference-Based RL, Intrinsic Motivation)
4. Policy Optimization (Offline RL, Imitation Learning, Hierarchical RL)
5. LLMs for Reasoning & Decision-Making (Causal Reasoning, Planning, Commonsense RL)
6. Exploration & Generalization (Zero-Shot RL, Generalization, Self-Supervised RL)
7. Multi-Agent RL (Emergent Communication, Coordination, Social RL)
8. Applications & Benchmarks (Autonomous Agents, Simulations, LLM-RL Benchmarks)
9. Tutorials & Courses (Lectures, Workshops)
10. Libraries & Implementations
11. Other Resources

Related Pages

Implements Principle

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment