
SY
Sen Yang, Ignacy Nieweglowski, Aviv Yaish
· 1 min read
ResearcharXiv cs.AI
FISSION: Label Augmentation for Bot Detection
arXiv:2609.26279v1 Announce Type: new
Abstract: Bot accounts and coordinated influence operations are often discovered via heuristic methods, leaving a dearth of reliable ground-truth labels for training detection systems. To address this challenge, we study a natural question: can we generate labels to assist in learning embeddings in which bots and accounts from the same coordinated operation are close? We present FISSION, a method to generate labels by splitting each account's activity into positively labeled sub-accounts. Given this label source, we train detection models which preserve behavioral regularities recurring across positive sub-accounts. We evaluate FISSION and show it outperforms prior methods in detecting Wikipedia sockpuppets and Twitter/X bots.
Original source
This story was published by arXiv cs.AI and written by Sen Yang, Ignacy Nieweglowski, Aviv Yaish. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


