Metraipolis › archive › dc_badd18580ddc

HeuFouFT: Task-Guided Metaheuristic Coordinate Search for Fourier Fine-Tuning — arXiv:2610.06437

Source
arXiv
Original
https://arxiv.org/abs/2610.06437
License (as archived)
CC0-1.0
Archived
2026-10-06T05:28:55.592Z
SHA-256 of the archived text
5746cfb7801a12aed6522c5bfb580aee847c041ece822aa1b56389e397d02514
Stable URL of this page
https://metraipolis.com/documento/dc_badd18580ddc

How to cite

HeuFouFT: Task-Guided Metaheuristic Coordinate Search for Fourier Fine-Tuning — arXiv:2610.06437 — arXiv, https://arxiv.org/abs/2610.06437 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_badd18580ddc (Metraipolis archive dc_badd18580ddc, sha256 5746cfb7801a12aed6522c5bfb580aee847c041ece822aa1b56389e397d02514)
BibTeX
@misc{dcbadd18580ddc,
  title = {HeuFouFT: Task-Guided Metaheuristic Coordinate Search for Fourier Fine-Tuning — arXiv:2610.06437},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.06437},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_badd18580ddc; sha256 5746cfb7801a12aed6522c5bfb580aee847c041ece822aa1b56389e397d02514},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.06437. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce Heuristic-Guided Fourier Fine-Tuning (HeuFouFT), a task-guided framework for selecting trainable frequency coordinates in Fourier fine-tuning. Existing uniform and Gaussian band-pass schemes allocate a limited spectral budget through fixed, task-agnostic rules. HeuFouFT instead searches for coordinates using downstream performance. A coarse intensity map from lightweight block-level probes initializes three metaheuristic optimizers: Genetic Algorithm with Simulated Annealing (GA-SA), Particle Swarm Optimization (PSO), and Cuckoo Search (CS). During search, a Random Forest filters each population so that only the top 30% of candidates proceed to proxy fine-tuning. On E2E with GPT-2-Medium, all three variants outperform random-uniform FourierFT, Gaussian band-pass FourierFT, and LoRA across five metrics. PSO further outperforms LoCA, the best-performing baseline, on four metrics while using 37.6% fewer trainable spectral coefficients. Once coordinates are selected, HeuFouFT requires only 15--18% FLOPs of Full FT. These results show that task-guided search allocates limited spectral capacity more effectively than fixed sampling. Our code is publicly available.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.06437 — 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 2026-10-06T05:24:20.462Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).