Metraipolis › archive › dc_111cd0990e8e

Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents — arXiv:2609.31076

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31076
License (as archived)
CC0-1.0
Archived
2026-10-08T01:38:32.388Z
SHA-256 of the archived text
a010b1423fc8ba0fb0c90eda52884f8f059a1557d16102c1fa0ef98b8f857313
Stable URL of this page
https://metraipolis.com/documento/dc_111cd0990e8e

How to cite

Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents — arXiv:2609.31076 — arXiv (en), https://arxiv.org/abs/2609.31076 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_111cd0990e8e (Metraipolis archive dc_111cd0990e8e, sha256 a010b1423fc8ba0fb0c90eda52884f8f059a1557d16102c1fa0ef98b8f857313)
BibTeX
@misc{dc111cd0990e8e,
  title = {Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents — arXiv:2609.31076},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31076},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_111cd0990e8e; sha256 a010b1423fc8ba0fb0c90eda52884f8f059a1557d16102c1fa0ef98b8f857313},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31076. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Language agents struggle to act and learn in environments that require long sequences of low-level actions. Code-based abstractions can make these agents more productive by letting them invoke reusable skills instead of repeatedly selecting individual actions. The code handles recurring local decisions, while the language model decides which skills to use and how to combine them. Yet abstractions are leaky, and situations beyond a skill's capabilities may require a return to primitive actions. Motivated by this tradeoff between productivity and flexibility, we systematically study how code-based action abstraction affects the performance, inference cost, and learning of language agents. We study this in NetHack, a challenging, long-horizon game environment, using CodeHack, our library of code-based skills with natural-language descriptions. We use this library to compare agents restricted to primitives with those using semantic skills alone or in combination with primitives. We evaluate these agents in three settings: zero-shot prompting, supervised fine-tuning, and reinforcement learning. Across a broad zero-shot evaluation on NetHack, we find that compared with primitives, skills nearly triple game progression, while reducing inference cost per episode by 86%. Combining skills with primitives retains much of this benefit while preserving a path back down to low-level actions. Finally, in RL, we find that skill-based agents learn significantly faster than agents acting on primitives, achieving a 7.2x larger average gain in dungeon level over the same training budget. These results show that a supplied skill library can improve performance, efficiency, and learning, while retaining primitives provides flexibility when the library is insufficient. We release CodeHack together with training and evaluation code.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.31076 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T01:12:14.335Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).