Summary of Raft: Realistic Attacks to Fool Text Detectors, by James Wang et al.
RAFT: Realistic Attacks to Fool Text Detectors
by James Wang, Ran Li, Junfeng Yang, Chengzhi Mao
First submitted to arxiv on: 4 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: 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 abstract presents a new attack method called RAFT, designed to compromise existing large language model (LLM) detectors by exploiting the transferability of LLM embeddings at the word-level. The authors show that their attacks can effectively compromise all detectors in the study across various domains by up to 99%, and are transferable across source models. Additionally, manual human evaluation studies reveal that examples generated by RAFT are realistic and indistinguishable from original human-written text. This work highlights the urgent need for more resilient detection mechanisms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper introduces a new attack method called RAFT, which targets existing LLM detectors. The authors show that their attacks can compromise all detectors across various domains by up to 99%. They also demonstrate that examples generated by RAFT are realistic and indistinguishable from human-written text. This highlights the need for more resilient detection mechanisms. |
Keywords
» Artificial intelligence » Large language model » Transferability
