TRACE-Bot: Detecting Emerging LLM-Driven Social Bots via Implicit Semantic Representations and AIGC-Enhanced Behavioral Patterns

2026-04-02

AI Summary

This work addresses the limitations of existing methods in detecting large language model (LLM)-driven social bots, which often rely on a single modality, lack sensitivity to generative AI (AIGC) patterns, and struggle to effectively integrate linguistic and behavioral cues. To overcome these challenges, we propose TRACE-Bot, a dual-channel unified framework that jointly models implicit semantic representations and AIGC-enhanced multidimensional behavioral features derived from user profiles, tweet content, and interaction dynamics to construct fine-grained user representations. By integrating pretrained language models with state-of-the-art AIGC detectors and employing a lightweight classification head, TRACE-Bot enables efficient and accurate bot identification. Evaluated on two public LLM-based social bot datasets, our method achieves accuracies of 98.46% and 97.50%, respectively, substantially outperforming current approaches and demonstrating strong robustness against sophisticated adversarial strategies.

Abstract

Large Language Model-driven (LLM-driven) social bots pose a growing threat to online discourse by generating human-like content that evades conventional detection. Existing methods suffer from limited detection accuracy due to overreliance on single-modality signals, insufficient sensitivity to the specific generative patterns of Artificial Intelligence-Generated Content (AIGC), and a failure to adequately model the interplay between linguistic patterns and behavioral dynamics. To address these limitations, we propose TRACE-Bot, a unified dual-channel framework that jointly models implicit semantic representations and AIGC-enhanced behavioral patterns. TRACE-Bot constructs fine-grained representations from heterogeneous sources, including personal information data, interaction behavior data and tweet data. A dual-channel architecture captures linguistic representations via a pretrained language model and behavioral irregularities via multidimensional activity features augmented with signals from state-of-the-art (SOTA) AIGC detectors. The fused representations are then classified through a lightweight prediction head. Experiments on two public LLM-driven social bot datasets demonstrate SOTA performance, achieving accuracies of 98.46% and 97.50%, respectively. The results further indicate strong robustness against advanced bot strategies, highlighting the effectiveness of jointly leveraging implicit semantic representations and AIGC-enhanced behavioral patterns for emerging LLM-driven social bot detection.

Problem

• LLM-driven social bots
• AIGC
• detection accuracy
• behavioral patterns
• semantic representations

Innovation

• LLM-driven social bots
• implicit semantic representations
• AIGC-enhanced behavioral patterns
• dual-channel framework
• AIGC detection