Metraipolis › archive › dc_c9e4decfe1ac
SkillLift: Learning Dense Rubrics from Sparse Oracles for Efficient Skill Evolution — arXiv:2609.15396
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.15396
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T22:47:58.450Z
- SHA-256 of the archived text
d1d5c6903530438f0cb5abf938e3759053dd64556939b0e777026d1517525fb3- Stable URL of this page
https://metraipolis.com/documento/dc_c9e4decfe1ac
How to cite
SkillLift: Learning Dense Rubrics from Sparse Oracles for Efficient Skill Evolution — arXiv:2609.15396 — arXiv (en), https://arxiv.org/abs/2609.15396 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c9e4decfe1ac (Metraipolis archive dc_c9e4decfe1ac, sha256 d1d5c6903530438f0cb5abf938e3759053dd64556939b0e777026d1517525fb3)
BibTeX
@misc{dcc9e4decfe1ac,
title = {SkillLift: Learning Dense Rubrics from Sparse Oracles for Efficient Skill Evolution — arXiv:2609.15396},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.15396},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c9e4decfe1ac; sha256 d1d5c6903530438f0cb5abf938e3759053dd64556939b0e777026d1517525fb3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15396. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM-based agents increasingly rely on persistent skills, i.e., reusable procedural prompts, to adapt without weight updates. Existing skill self-evolution methods directly revise skill text based on execution feedback, but each oracle evaluation requires a full agent rollout, creating a supervision bottleneck that confines search to failure-patching updates. Our key insight is that ranking is a smoother supervision target than absolute outcome regression: identifying which skill is better requires fewer oracle evaluations than predicting exact scores. Building on this insight, we propose SkillLift, which decouples skill search from oracle cost by learning an oracle-aligned rubric as a structured evaluation space. We formalize this as a bilevel optimization problem solved via alternating optimization: an inner loop uses the frozen rubric as a cheap surrogate to guide skill revision at no oracle cost, while an outer loop invokes a small number of oracle rollouts to re-align the rubric via rank correlation, amortizing oracle cost and stabilizing text-space updates. Experiments on complex agent task benchmarks show that our method outperforms existing auto-skill methods with 40--70\% less token cost compared to frontier evolving methods. Codes are available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.15396 — 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 2026-10-15T22:29:03.310Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).