[0scarJ1ang]

Adds Agent Priors-guided Policy Learning (APPL) to Brain Models, next to the skill-centric planning entries (ClevrSkills, RoboMatrix).

• Paper: https://arxiv.org/abs/2609.35690 (arXiv v1, September 28, 2026)
• Project page: https://agentics-robotics.github.io/APPL/
• Code: https://github.com/Agentics-robotics/Agent-Priors-guided-Policy-Learning

APPL uses a language-model agent as the high-level "brain" over learned skill policies. A construction agent segments demonstrations into skills and trains and verifies one policy per proposed structural prior. A runtime agent selects and composes these policies through interfaces that state each prior's applicability. Marked [x] because the official code is public; Project Link points to the code repository.

Checklist:

• I checked for duplicates.
• I verified the paper link.
• I verified the project/code/dataset link.
• I used the correct year.
• I ran python scripts/check_readme.py (passes).

Disclosure: submitted by the paper authors.

🤖 Generated with Claude Code