Researchers Released NOMOS Secure Retrieval Protocol

The new NOMOS protocol significantly improves efficiency in encrypted database ranking tasks.

Updated on Sept. 30, 2026 in Cybersecurity

Isometric editorial illustration featuring a dense, structured grid of matte crystalline pillars in teal and cream tones.
Researchers have released the NOMOS protocol, a new secure data retrieval method designed to reduce latency in homomorphic encryption systems. AI Illustration. Upload story photo >

Researchers have released the NOMOS protocol, a new secure k-nearest neighbor method that addresses significant latency bottlenecks in homomorphic encryption systems. The protocol enables high-accuracy data retrieval while maintaining strict security standards.

Why it matters

The system addresses the primary challenge of ranking and top-k extraction within homomorphic encryption, providing a pathway for faster secure data processing. This development enables more practical applications for encrypted databases where performance has traditionally lagged.

NOMOS utilizes a gap-amplified sign approximation method alongside offline k-means preprocessing to form candidate sets. It achieves latency reductions of nearly 8 times over Mazzone-style benchmarks and up to 828 times compared to Engorgio.

The details

NOMOS integrates slot-index alignment and ReLU-based top-k extraction to handle complex queries efficiently. By focusing precision on near-zero distance gaps, the protocol ensures high performance without sacrificing the retrieval accuracy required for secure environments.

Timeline

  1. September 29, 2026: The research paper detailing the NOMOS protocol was released.

The Tech Race

NOMOS marks a significant performance evolution for systems built upon the foundation of the CKKS homomorphic encryption scheme. By solving long-standing speed limitations, it positions itself against older architectures like Engorgio that have struggled with high query latency.

As this protocol reaches enterprise software providers, users may experience faster, more secure interactions with cloud-based services that process sensitive information. It effectively lowers the performance tax currently associated with using end-to-end encrypted databases.

The takeaway

This breakthrough demonstrates that homomorphic encryption can overcome its reputation for slow speeds through specialized, targeted approximation methods. Developers should look to these techniques to balance high-level security with the responsiveness required by modern consumer applications.

Further reading

For broader context on current cryptographic developments, visit our Cybersecurity section.

More information

Read the complete NOMOS encrypted kNN protocol research paper for further technical specifications.