Summary of Detecting Training Data Of Large Language Models Via Expectation Maximization, by Gyuwan Kim et al.
Detecting Training Data of Large Language Models via Expectation Maximization
by Gyuwan Kim, Yang Li, Evangelia Spiliopoulou, Jie Ma, Miguel Ballesteros, William Yang Wang
First submitted to arxiv on: 10 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); 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 This paper tackles the problem of membership inference attacks (MIAs) on large language models (LLMs), which aim to determine whether a specific instance was part of a target model’s training data. The authors propose EM-MIA, an expectation-maximization algorithm-based MIA method that iteratively refines membership scores and prefix scores. This approach achieves state-of-the-art results on the WikiMIA dataset. To further evaluate EM-MIA, the paper presents OLMoMIA, a benchmark built from OLMo resources that allows for controlling the difficulty of MIA tasks with varying degrees of overlap between training and test data distributions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about how to figure out if a certain piece of text was used to train a big language model. This is important because it helps us understand what’s going on inside these models and make sure they’re not using bad information or breaking rules. The authors created a new way to do this, called EM-MIA, which works really well on some tests. They also made a special tool, called OLMoMIA, that lets them test their method in different situations. |
Keywords
» Artificial intelligence » Inference » Language model
