Metraipolis › archive › dc_e3d35ebf7157
MERA: Model Evolution and Routing with Skill Adaptation for Agentic Systems at Scale — arXiv:2608.10333
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.10333
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T06:19:18.570Z
- SHA-256 of the archived text
d875eb9e239dad0ed99b5eb5cb2966ee761435873e2a9e2829e63db64dc1065d- Stable URL of this page
https://metraipolis.com/documento/dc_e3d35ebf7157
How to cite
MERA: Model Evolution and Routing with Skill Adaptation for Agentic Systems at Scale — arXiv:2608.10333 — arXiv (en), https://arxiv.org/abs/2608.10333 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e3d35ebf7157 (Metraipolis archive dc_e3d35ebf7157, sha256 d875eb9e239dad0ed99b5eb5cb2966ee761435873e2a9e2829e63db64dc1065d)
BibTeX
@misc{dce3d35ebf7157,
title = {MERA: Model Evolution and Routing with Skill Adaptation for Agentic Systems at Scale — arXiv:2608.10333},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.10333},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e3d35ebf7157; sha256 d875eb9e239dad0ed99b5eb5cb2966ee761435873e2a9e2829e63db64dc1065d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10333. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM agents execute heterogeneous sequences of model calls within a single task: some invocations require careful reasoning, while others are structured steps such as formatting or tool-argument construction. Prior routing methods exploit this asymmetry by assigning easy invocations to a cheaper small model and difficult ones to a large model. Such policies reduce inference cost, but they leave the small model's capability unchanged, so attainable savings remain bounded by the work the student can already solve. MERA instead improves the small model itself, using a single model invocation as the unit of adaptation. In each cycle, MERA replays failed student invocations to obtain execution-verified teacher demonstrations, distills recurring procedures into an iteratively updated SkillBook, and fine-tunes a student LoRA adapter via supervised learning and optional GRPO. Routing serves as supporting machinery for deployment: the improved student is served behind a cost-calibrated router with verifier-backed fallback, and a candidate SkillBook, adapter, or router is admitted only when joint replay preserves task quality. Empirically, four-cycle adaptation raises Qwen2.5-Coder-1.5B from 28.7% to 49.7% pass on held-out HumanEval+MBPP. Under verifier-backed fallback, the deployed policy retains 88.3% pass at 60.8% of always-Luna cost. On TAU-2, a fine-tuned Qwen3.5-2B improves from 14/35 to 18/35 and matches an unadapted 4B model. These results indicate that verifier-backed multi-cycle adaptation can increase small-model capability, rather than only routing around a fixed student. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.10333 — 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 2026-10-19T06:11:45.072Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).