tool arXiv Apr 21, 2026 · 4w ago
Zhao Wang, Max Xiong, Jianxun Lian et al. · Renmin University of China · Duke University +1 more
Reasoning-driven AI text detector using reinforcement learning to generate interpretable explanations before classification across diverse LLM sources
Output Integrity Attack nlp
The rapid advancement and widespread adoption of Large Language Models (LLMs) have elevated the need for reliable AI-generated content (AIGC) detection, which remains challenging as models evolve. We introduce AIGC-text-bank, a comprehensive multi-domain dataset with diverse LLM sources and authorship scenarios, and propose REVEAL, a detection framework that generates interpretable reasoning chains before classification. Our approach uses a two-stage training strategy: supervised fine-tuning to establish reasoning capabilities, followed by reinforcement learning to improve accuracy, improve logical consistency, and reduce hallucinations. Extensive experiments show that REVEAL achieves state-of-the-art performance across multiple benchmarks, offering a robust and transparent solution for AIGC detection. The project is open-source at https://aka.ms/reveal
llm transformer Renmin University of China · Duke University · Microsoft Research
ML Security Papers