AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Group Testing for Accurate and Efficient Range-Based Near Neighbor Search for Plagiarism Detection

Harsh Shah, Kashish Mittal, Ajit Rajwade

ECCV 2024 · Learning theory and algorithms › Prediction and decision losses · approach: Approximate nearest neighbor indexing · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use binary-splitting group testing instead of exhaustive search for range NN searchOur-Sum · MIRFLICKR, rho=0.8mean query time (ms): 1222 -> 54.0Improved
use binary-splitting group testing instead of exhaustive searchOur-Sum · IMDB-Wiki, rho=0.8mean query time (ms): 800 -> 98.8Improved
sum pooling instead of element-wise max pooling (ablation)binary splitting · MIRFLICKR, rho=0.8mean query time (ms): 81 -> 54.0Improved
sum pooling instead of element-wise max pooling (ablation)binary splitting · ImageNet, rho=0.8dot products per query: 12339 (sum) vs 5713 (max)Hurt
use Our-Sum instead of Flinng / Flann-K approximate group testingrange NN search · MIRFLICKR, rho=0.8mean recall: 0.268 -> 1.0 (Flinng)Improved
use Our-Sum instead of Hnsw in streaming insertionstreaming setting · ImageNet, streamingmean insertion time (ms): 236 -> 0.28Improved
use Scann instead of Our-Sumrange NN search · MIRFLICKR, rho=0.8mean query time (ms): 54.0 -> 23 (recall 1.0 -> 0.996)Improved
use Hnsw instead of Our-Sum on IMDB-Wikirange NN search · IMDB-Wiki, rho=0.8mean query time (ms): 98.8 -> 1076Hurt
apply compressed-sensing expander recovery to approximate NN searchgroup testing NN search · simulationsrecovery error boundHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy