FROMANNUAL REVIEWS

Lecture image placeholder

Premium content

Access to this content requires a subscription. You must be a premium user to view this content.

Monthly subscription - $9.99Pay per view - $4.99Access through your institutionLogin with Underline account
Need help?
Contact us
Lecture placeholder background

IRPS 2026 Main Conference & YIR

•

March 25, 2026

•

Tucson, United States

keywords:

approximate nearest neighbor search

3d-nand

neuromorphic computing

This paper demonstrates a highly reliable 3D-NAND-based billion-scale approximate nearest neighbor search (ANNS) accelerator, which can evaluate the Manhatten (L1) distance between the query and multi-billions of instances directly in the SSD. Three approaches (1. “Partial-ON” string-select line (SSL) input, 2. Thermometer code data-encoding scheme  , and 3. On-demand SSL re-trimming) are discussed to improve the accuracy and stability. Proposed computational SSD can maintain high recall rate for the SIFT-1B test even after 200℃ baking.

Downloads

SlidesPaperTranscript English (automatic)
access premium content

Similar lecture

Reliability Assessment for an In-3D-NAND Approximate Searching Solution
technical paper

Reliability Assessment for an In-3D-NAND Approximate Searching Solution

IRPS 2024 Main Conference

Po-Hao Tseng
+5
Po-Hao Tseng and 7 other authors

17 April 2024

Stay up to date with the latest Underline news!

Select topic of interest (you can select more than one)

PRESENTATIONS

  • All Presentations
  • For Librarians
  • Resource Center
  • Free Trial
Underline Science, Inc.
1216 Broadway, 2nd Floor, New York, NY 10001, USA

© 2026 Underline - All rights reserved

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.