Metraipolis › archive › dc_0ea486b22e40

Fiona: Accelerating FHE Inference with Packing-Aware Ternary Weights — arXiv:2609.35352

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35352
License (as archived)
CC0-1.0
Archived
2026-10-07T22:39:05.557Z
SHA-256 of the archived text
9482a7050829706447e55665d97ecfbfd3ece031f91437701b9ded20dd70b85f
Stable URL of this page
https://metraipolis.com/documento/dc_0ea486b22e40

How to cite

Fiona: Accelerating FHE Inference with Packing-Aware Ternary Weights — arXiv:2609.35352 — arXiv (en), https://arxiv.org/abs/2609.35352 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0ea486b22e40 (Metraipolis archive dc_0ea486b22e40, sha256 9482a7050829706447e55665d97ecfbfd3ece031f91437701b9ded20dd70b85f)
BibTeX
@misc{dc0ea486b22e40,
  title = {Fiona: Accelerating FHE Inference with Packing-Aware Ternary Weights — arXiv:2609.35352},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35352},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0ea486b22e40; sha256 9482a7050829706447e55665d97ecfbfd3ece031f91437701b9ded20dd70b85f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35352. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fully homomorphic encryption (FHE) enables neural network inference directly on encrypted inputs, but it remains orders of magnitude slower than plaintext in- ference. Applying the server's plaintext weights to encrypted activations involves plaintext-ciphertext multiplications (PMult) and accounts for more than half of inference time in recent systems. Ternary quantization can replace these multipli- cations with additions and subtractions, but the savings rarely materialize under packed execution. A single PMult applies a weight group fixed by the packing layout and can be avoided only when all its weights share the same ternary value. Ternarizing all groups, however, largely degrades accuracy. We present FIONA, an offline optimizer that selectively ternarizes weights within a given packing layout based on the estimated effect of ternary conversion on the model's performance. FIONA encourages a shared ternary value within each weight group and retains full-precision weights for sensitive groups, so ternar- ized and full-precision paths coexist within a layer. It then compiles these hybrid operators exactly, applying common scaling factors once to accumulated inputs and reusing sums across outputs. Weight ternarization can also narrow the input ranges of downstream polynomials. FIONA fits lower-degree replacements under a cumulative accuracy budget, reducing multiplicative depth and bootstrapping. On VGG11, ViT, and BERT, FIONA reduces PMult operations by 53.4-79.5% and accelerates end-to-end encrypted inference by 2.38x, 1.68x, and 1.84x, re- spectively, with less than 1% accuracy loss across all three models.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.35352 — 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 2026-10-07T22:14:28.429Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).