Metraipolis › archive › dc_18df7428fdef
ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization — arXiv:2610.00906
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00906
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T11:54:47.106Z
- SHA-256 of the archived text
73d7b61ed349cf9e9a7de7cc225c7003693eafc662631d8db44b61f98044527b- Stable URL of this page
https://metraipolis.com/documento/dc_18df7428fdef
How to cite
ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization — arXiv:2610.00906 — arXiv (en), https://arxiv.org/abs/2610.00906 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_18df7428fdef (Metraipolis archive dc_18df7428fdef, sha256 73d7b61ed349cf9e9a7de7cc225c7003693eafc662631d8db44b61f98044527b)
BibTeX
@misc{dc18df7428fdef,
title = {ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization — arXiv:2610.00906},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00906},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_18df7428fdef; sha256 73d7b61ed349cf9e9a7de7cc225c7003693eafc662631d8db44b61f98044527b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00906. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Automated harness optimization can substantially improve LLM agents by iteratively updating their prompts, tool interfaces, and control logic from execution feedback. However, existing methods primarily optimize how the harness is updated while largely fixing which training scenarios generate the feedback that drives those updates. As the harness evolves, the scenarios most useful for further optimization can change, suggesting that the training curriculum itself should adapt alongside the harness. We formulate this missing dimension of harness optimization as an automated curriculum learning problem and introduce ActiveSaddler. ActiveSaddler models the evolving curriculum as a non-stationary bandit with dynamically instantiated optimization targets. It abstracts recurring failures into reusable failure-pattern arms, estimates the potential learning progress from further targeting each pattern, and adaptively balances revisiting known weaknesses with exploring unseen scenarios for new ones. Optimization outcomes continually update both the set of discovered failure patterns and their priorities, allowing the curriculum to co-evolve with the harness. Experiments on GAIA2 and Terminal-Bench 2.0 show that ActiveSaddler consistently discovers stronger harnesses, improving test Pass@1 by 4.4 and 7.5 percentage points over the same harness optimizer using a scenario order fixed before optimization, respectively. Ablations further show that these gains depend on dynamically constructing optimization targets, estimating their evolving utility, and balancing continued optimization with new failure discovery. Together, these results establish automated curriculum learning as a new crucial optimization dimension for harness optimization. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.MA). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2610.00906 — 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 2026-10-06T10:17:30.567Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).