Salvato in:
Dettagli Bibliografici
Autori principali: Thomas, Rahul, Zahran, Louai, Choi, Erica, Potti, Akilesh, Goldblum, Micah, Pal, Arka
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2507.05228
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918085444763648
author Thomas, Rahul
Zahran, Louai
Choi, Erica
Potti, Akilesh
Goldblum, Micah
Pal, Arka
author_facet Thomas, Rahul
Zahran, Louai
Choi, Erica
Potti, Akilesh
Goldblum, Micah
Pal, Arka
contents As LLMs continue to increase in parameter size, the computational resources required to run them are available to fewer parties. Therefore, third-party inference services -- where LLMs are hosted by third parties with significant computational resources -- are becoming increasingly popular. However, third party inference raises critical concerns about user data privacy. To mitigate these risks, privacy researchers have developed provably secure schemes for third-party inference, such as Secure Multi-Party Computation (SMPC). However, SMPC protocols have significant computational and communication overhead, and do not scale to large models. In this work, we propose a new multi-party inference protocol, Cascade, that avoids these punitive costs by leveraging sharding in the sequence dimension to maintain privacy, trading off cryptographic privacy guarantees for increased performance and scalability. We demonstrate that Cascade is resistant to a generalization of a recent attack that is highly effective against other statistical privacy schemes, and that it is further resistant to learning-based attacks. As Cascade is orders of magnitude faster than existing schemes, our findings offer practical solutions for secure deployment of modern state-of-the-art LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2507_05228
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cascade: Token-Sharded Private LLM Inference
Thomas, Rahul
Zahran, Louai
Choi, Erica
Potti, Akilesh
Goldblum, Micah
Pal, Arka
Machine Learning
Cryptography and Security
As LLMs continue to increase in parameter size, the computational resources required to run them are available to fewer parties. Therefore, third-party inference services -- where LLMs are hosted by third parties with significant computational resources -- are becoming increasingly popular. However, third party inference raises critical concerns about user data privacy. To mitigate these risks, privacy researchers have developed provably secure schemes for third-party inference, such as Secure Multi-Party Computation (SMPC). However, SMPC protocols have significant computational and communication overhead, and do not scale to large models. In this work, we propose a new multi-party inference protocol, Cascade, that avoids these punitive costs by leveraging sharding in the sequence dimension to maintain privacy, trading off cryptographic privacy guarantees for increased performance and scalability. We demonstrate that Cascade is resistant to a generalization of a recent attack that is highly effective against other statistical privacy schemes, and that it is further resistant to learning-based attacks. As Cascade is orders of magnitude faster than existing schemes, our findings offer practical solutions for secure deployment of modern state-of-the-art LLMs.
title Cascade: Token-Sharded Private LLM Inference
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2507.05228