Metraipolis › archive › dc_80e01a3cf37b
Beyond FLOPs: Energy-Aware Knowledge Distillation for Sustainable LLMs on Code-Related Task — arXiv:2608.17515
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.17515
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T14:33:32.060Z
- SHA-256 of the archived text
f5451d2d407cddaf01287a187a34477501c07dd2f98d2d7da0529e795ad8f2b8- Stable URL of this page
https://metraipolis.com/documento/dc_80e01a3cf37b
How to cite
Beyond FLOPs: Energy-Aware Knowledge Distillation for Sustainable LLMs on Code-Related Task — arXiv:2608.17515 — arXiv (en), https://arxiv.org/abs/2608.17515 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80e01a3cf37b (Metraipolis archive dc_80e01a3cf37b, sha256 f5451d2d407cddaf01287a187a34477501c07dd2f98d2d7da0529e795ad8f2b8)
BibTeX
@misc{dc80e01a3cf37b,
title = {Beyond FLOPs: Energy-Aware Knowledge Distillation for Sustainable LLMs on Code-Related Task — arXiv:2608.17515},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.17515},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80e01a3cf37b; sha256 f5451d2d407cddaf01287a187a34477501c07dd2f98d2d7da0529e795ad8f2b8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.17515. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Background: Large Language Models (LLMs) are increasingly being applied to Software Engineering (SE) tasks, achieving high accuracy across problems such as clone detection, vulnerability prediction, and code summarization. However, their high computational demands and energy consumption raise sustainability concerns and hinder their use on consumer hardware and resource-constrained platforms. A common way to report the computational cost of an LLM in the literature and industry is to use the number of Floating Point Operations (FLOPs) required to perform a pass over the network. Aims: This paper investigates the implications of energy-aware knowledge distillation for SE, aiming to improve model efficiency while maintaining performance and to determine whether FLOPs is a reliable energy-aware metric. Method: We conduct a controlled experiment using Morph, a Many-Objective Optimization-based distillation methodology, to empirically examine whether FLOPs accurately reflect energy consumption in Clone Detection and Vulnerability Prediction tasks. We extend this methodology to include energy-surrogate models that directly estimate CPU and GPU energy consumption during optimization, and we apply Morph to generative tasks using CodeT5+ for code summarization. Results: Our results show that FLOPs is not always a reliable indicator of energy consumption, and better results can be achieved by using energy-surrogate models. Distilled student models can reduce inference energy consumption by up to 90\% and memory usage by 86\%, with only modest accuracy trade-offs. Conclusions: Energy-aware knowledge distillation when guided by direct energy surrogates rather than FLOPs can improve the energy consumption, sustainability, and deployability of LLMs for SE applications, enabling efficient models on consumer hardware. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-19 in cs.AI). Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.17515 — 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 2026-10-18T14:27:54.429Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-19?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).