Summary of Detective: Detecting Ai-generated Text Via Multi-level Contrastive Learning, by Xun Guo et al.
DeTeCtive: Detecting AI-generated Text via Multi-Level Contrastive Learning
by Xun Guo, Shan Zhang, Yongxin He, Ting Zhang, Wanquan Feng, Haibin Huang, Chongyang Ma
First submitted to arxiv on: 28 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach to detecting AI-generated text, which is currently limited by manual feature crafting and supervised binary classification paradigms. The authors argue that the key lies in distinguishing writing styles of different authors rather than classifying text as human-written or AI-generated. They introduce DeTeCtive, a multi-task auxiliary framework combining contrastive learning with dense information retrieval for AI-generated text detection. The method is compatible with various text encoders and achieves state-of-the-art results on multiple benchmarks. In out-of-distribution zero-shot evaluation, it outperforms existing approaches by a large margin. The paper also demonstrates Training-Free Incremental Adaptation capability towards out-of-distribution data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Detecting AI-generated text is important for ensuring the safe application of large language models (LLMs). Current methods are limited and often don’t work well with out-of-distribution data. This paper proposes a new approach called DeTeCtive that can detect AI-generated text by identifying writing styles. It combines contrastive learning and information retrieval to improve detection accuracy. The method works well even when it’s tested on data it hasn’t seen before, which is important for real-world applications. |
Keywords
» Artificial intelligence » Classification » Multi task » Supervised » Zero shot
