SOGPTSpotter: Detecting ChatGPT-Generated Answers on Stack Overflow

TL;DR

SOGPTSpotter is a novel Siamese Neural Network-based method that effectively detects ChatGPT-generated answers on Stack Overflow, outperforming existing baselines and aiding moderation efforts.

Contribution

Introduces SOGPTSpotter, a Siamese Neural Network approach utilizing BigBird and Triplet loss for accurate detection of AI-generated answers on Stack Overflow.

Findings

Outperforms baseline detection methods like GPTZero and DetectGPT.

Effective in real-world moderation by identifying ChatGPT-suspected answers.

Robust against adversarial attacks and generalizes across domains.

Abstract

Stack Overflow is a popular Q&A platform where users ask technical questions and receive answers from a community of experts. Recently, there has been a significant increase in the number of answers generated by ChatGPT, which can lead to incorrect and unreliable information being posted on the site. While Stack Overflow has banned such AI-generated content, detecting whether a post is ChatGPT-generated remains a challenging task. We introduce a novel approach, SOGPTSpotter, that employs Siamese Neural Networks, leveraging the BigBird model and the Triplet loss, to detect ChatGPT-generated answers on Stack Overflow. We use triplets of human answers, reference answers, and ChatGPT answers. Our empirical evaluation reveals that our approach outperforms well-established baselines like GPTZero, DetectGPT, GLTR, BERT, RoBERTa, and GPT-2 in identifying ChatGPT-synthesized Stack Overflow responses. We also conducted an ablation study to show the effectiveness of our model. Additional experiments were conducted to assess various factors, including the impact of text length, the model's robustness against adversarial attacks, and its generalization capabilities across different domains and large language models. We also conducted a real-world case study on Stack Overflow. Using our tool's recommendations, Stack Overflow moderators were able to identify and take down ChatGPT-suspected generated answers, demonstrating the practical applicability and effectiveness of our approach.