AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

EMBERSim: A Large-Scale Databank for Boosting Similarity Search in Malware Analysis

Dragos Georgian Corlatescu, Alexandru Dinu, Mihaela Gaman, Paul Sumedrea

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 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 XGBoost leaf similarity instead of ssdeepsimilarity search · EMBER label homogeneity, K=100, allmean same-label hits: 40.42 -> 94.53Improved
use leaf similarity instead of ssdeepsimilarity search · EMBER label homogeneity, K=10, allmean same-label hits: 5.43 -> 9.74Improved
use leaf similarity instead of ssdeepsimilarity search · EMBER label homogeneity, K=100, maliciousmean same-label hits: 61 -> 96.28Improved
use leaf similarity instead of ssdeepsimilarity search · EMBER label homogeneity, K=100, benignmean same-label hits: 19.84 -> 92.80Improved
increase top-K from 1 to 100leaf similarity · EMBER MAP, counterfactual, maliciousmean average precision: 0.960 -> 0.783Hurt
evaluate malicious instead of benign queriesleaf similarity · EMBER relevance EM, counterfactual, K=100relevance: 0.933 -> 0.641Hurt
use IoU instead of exact match as relevance functionleaf similarity · EMBER counterfactual, malicious, K=100relevance: 0.641 -> 0.759Improved
use NES instead of IoU as relevance functionleaf similarity · EMBER counterfactual, malicious, K=100relevance: 0.759 -> 0.752No clear effect
counterfactual (out-of-time) evaluation instead of unsupervised labellingleaf similarity · EMBER MAP, malicious, K=1mean average precision: 0.716 -> 0.960Improved
apply leaf similarity to DREBIN instead of EMBERleaf similarity · label homogeneity, K=100, allmean same-label hits: 94.53 -> 98.30Improved
DREBIN malicious queries at K=100leaf similarity · DREBIN label homogeneitymean same-label hits: 98.30 -> 71.21Hurt

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