ForensicsData: a Digital Forensics Dataset for Large Language Models

Youssef Chakir 1 , Iyad Lahsen-Cherif 1 · 2025

Read ForensicsData: a Digital Forensics Dataset for Large Language Models by Youssef Chakir 1 , Iyad Lahsen-Cherif 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-05331.txt.

Description

Digital forensic investigators need realistic practice data that privacy rules rarely allow, and Youssef Chakir with Iyad Lahsen-Cherif answer with a dataset in a 2025 paper. ForensicsData is built from actual malware analysis reports and organised as question, context, and answer triples.

The set holds more than five thousand such triples. A workflow extracts structured facts from reports, uses large language models to reshape them into the triple format, then checks quality with a dedicated evaluation step. Among models tested, one commercial system aligned best with forensic wording. The authors aim to support reproducible experiments with language models on forensic tasks and to widen collaboration in a field short on shared data.

Catalog record

CREATORYoussef Chakir 1 , Iyad Lahsen-Cherif 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-05331

Browse the full library · About the Tower of Babel