Metraipolis › archive › dc_0389ca8ca351

HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation — arXiv:2609.01730

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01730
License (as archived)
CC0-1.0
Archived
2026-10-17T03:26:48.904Z
SHA-256 of the archived text
1539f13dc2df6c5b203175f4dc8af3bf92ca69de00040101743f19fda91c88aa
Stable URL of this page
https://metraipolis.com/documento/dc_0389ca8ca351

How to cite

HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation — arXiv:2609.01730 — arXiv (en), https://arxiv.org/abs/2609.01730 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0389ca8ca351 (Metraipolis archive dc_0389ca8ca351, sha256 1539f13dc2df6c5b203175f4dc8af3bf92ca69de00040101743f19fda91c88aa)
BibTeX
@misc{dc0389ca8ca351,
  title = {HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation — arXiv:2609.01730},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01730},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0389ca8ca351; sha256 1539f13dc2df6c5b203175f4dc8af3bf92ca69de00040101743f19fda91c88aa},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01730. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fully homomorphic encryption (FHE) allows a server to run a language model directly on encrypted user prompts, but current approaches remain prohibitively slow. Ciphertexts natively support only addition, multiplication, and rotation, and multiplications may be composed only to a bounded depth before a costly bootstrapping operation is required to continue. Every nonlinearity must therefore be approximated by an iterative method; each iteration increasing the number of multiplications. A higher iteration count buys precision but exhausts the available depth more frequently and thus triggers more bootstraps, which dominate latency. We introduce Homomorphic Encryption-Aware Training (HEAT), a fine-tuning method that makes the per-nonlinearity iteration counts learnable, enabling them and the model weights to co-adapt during training. HEAT optimizes iterations with respect to the task objective, allowing the model to adapt to approximation errors encountered during inference without architectural changes or retraining from scratch. We further relate iteration count to quantization bit width and bound, at fixed weights, the gap between our objective and quantization-aware training. On encrypted GPT-2 decoding, HEAT reduces iterations by $3.1\times$, bootstraps by $1.6\times$, and end-to-end latency by $1.4\times$, while improving decode agreement over the calibrated encrypted baseline.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.AI).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.01730 — 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 2026-10-17T03:13:12.475Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).