SOREL-20M: A Large Scale Benchmark Dataset for Malicious PE Detection

About

In this paper we describe the SOREL-20M (Sophos/ReversingLabs-20 Million) dataset: a large-scale dataset consisting of nearly 20 million files with pre-extracted features and metadata, high-quality labels derived from multiple sources, information about vendor detections of the malware samples at the time of collection, and additional ``tags'' related to each malware sample to serve as additional targets. In addition to features and metadata, we also provide approximately 10 million ``disarmed'' malware samples -- samples with both the optional\_headers.subsystem and file\_header.machine flags set to zero -- that may be used for further exploration of features and detection strategies. We also provide Python code to interact with the data and features, as well as baseline neural network and gradient boosted decision tree models and their results, with full training and evaluation code, to serve as a starting point for further experimentation.

Richard Harang, Ethan M. Rudd · 2020

Related benchmarks

Task	Dataset	Result
Malware Detection	Set A real malware with constant padding attack 2020	Accuracy69.5	6
Malware Detection	EMBER (test)	Correct Allocations Count—	6
Malware Detection	Set B real adversarial malware with Greedy Attack 2021	Accuracy74.9	5