Reasoning-Aware AIGC Detection via Alignment and Reinforcement

2026-04-21

AI Summary

This work addresses the growing challenge of detecting machine-generated text as large language models produce increasingly human-like outputs. The authors propose REVEAL, a novel framework that enhances detection generalization and robustness through a two-stage strategy: first generating interpretable reasoning chains and then performing classification. To support this approach, they introduce AIGC-text-bank, a multi-source, multi-domain benchmark dataset for AI-generated content detection. Notably, this is the first effort to incorporate interpretable reasoning chains into AIGC detection, leveraging supervised fine-tuning and reinforcement learning to improve logical consistency and mitigate hallucinations. Experimental results demonstrate that REVEAL significantly outperforms existing methods across multiple benchmarks. The code has been publicly released, establishing a reliable and transparent new paradigm for AIGC detection.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Large Vision Models

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

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

Problem

• AIGC detection
• Large Language Models
• AI-generated content
• reasoning-aware detection

Innovation

• reasoning-aware detection
• reinforcement learning
• interpretable reasoning chains
• AIGC-text-bank
• two-stage training