What Does the Bot Say? Opportunities and Risks of Large Language Models in Social Media Bot Detection

TL;DR

This paper explores how large language models can enhance social media bot detection through novel frameworks, but also pose risks by enabling manipulation that can undermine detection accuracy and system reliability.

Contribution

It introduces a mixture-of-heterogeneous-experts framework for LLM-based bot detection and analyzes the risks of LLM-guided manipulation to evade detection.

Findings

Instruction tuning with 1,000 examples improves detection accuracy by up to 9.1%.

LLM-guided manipulation can reduce detection performance by up to 29.6%.

Manipulation strategies harm the calibration and reliability of detection systems.

Abstract

Social media bot detection has always been an arms race between advancements in machine learning bot detectors and adversarial bot strategies to evade detection. In this work, we bring the arms race to the next level by investigating the opportunities and risks of state-of-the-art large language models (LLMs) in social bot detection. To investigate the opportunities, we design novel LLM-based bot detectors by proposing a mixture-of-heterogeneous-experts framework to divide and conquer diverse user information modalities. To illuminate the risks, we explore the possibility of LLM-guided manipulation of user textual and structured information to evade detection. Extensive experiments with three LLMs on two datasets demonstrate that instruction tuning on merely 1,000 annotated examples produces specialized LLMs that outperform state-of-the-art baselines by up to 9.1% on both datasets, while LLM-guided manipulation strategies could significantly bring down the performance of existing bot detectors by up to 29.6% and harm the calibration and reliability of bot detection systems.