Metraipolis › archive › dc_04dce659e075
SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent — arXiv:2608.07449
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.07449
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T09:47:15.467Z
- SHA-256 of the archived text
a82094e910bcabd6a72e7c7053e31510c8aea35631dd3d4ca58d2c787dec19bb- Stable URL of this page
https://metraipolis.com/documento/dc_04dce659e075
How to cite
SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent — arXiv:2608.07449 — arXiv (en), https://arxiv.org/abs/2608.07449 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_04dce659e075 (Metraipolis archive dc_04dce659e075, sha256 a82094e910bcabd6a72e7c7053e31510c8aea35631dd3d4ca58d2c787dec19bb)
BibTeX
@misc{dc04dce659e075,
title = {SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent — arXiv:2608.07449},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.07449},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_04dce659e075; sha256 a82094e910bcabd6a72e7c7053e31510c8aea35631dd3d4ca58d2c787dec19bb},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.07449. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM agents increasingly adapt to recurring tasks by accumulating procedural knowledge in skills. These skills are lightweight, reusable textual artifacts that are loaded into the agent's context without weight updates. Recent methods refine skills through iterative task execution, failure diagnosis, and trajectory-guided text-space updates. However, existing frameworks lack explicit diagnosis--outcome feedback and treat deletion as a generic edit operation rather than a dedicated mechanism for consolidating accumulated knowledge. We introduce SkillProx, a proximal-gradient-inspired forward--backward framework that couples closed-loop diagnostic evolution with utility-aware proximal refinement. Motivated by a composite objective balancing task loss and skill complexity, the forward stage re-executes diagnosis-driven edits on the same task batch, rolls back regressions, and feeds measured outcomes into subsequent diagnoses. The backward stage decomposes the resulting skill into auditable knowledge units, estimates their contributions using a frozen leave-one-out utility audit, and applies validation-gated consolidation, demotion, or removal. Experiments on in-distribution and out-of-distribution benchmarks across multiple backbone LLMs show that SkillProx improves average accuracy by 3.0 percentage points over the strongest gradient-based baseline. Component ablations demonstrate the complementary effects of closed-loop diagnosis and proximal refinement. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-10 in cs.CL). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.07449 — 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 2026-10-19T09:46:37.278Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).