文本和圖像都被洩露了！對多模態LLM的數據污染進行系統分析

Both Text and Images Leaked! A Systematic Analysis of Multimodal LLM Data Contamination

November 6, 2024

作者: Dingjie Song, Sicheng Lai, Shunian Chen, Lichao Sun, Benyou Wang

cs.AI

摘要

多模式大型語言模型（MLLMs）的快速發展在各種多模式基準測試中展現出卓越的性能。然而，在訓練過程中出現的數據污染問題對性能評估和比較構成挑戰。儘管存在許多用於檢測大型語言模型（LLMs）中數據集污染的方法，但由於多模式和多個訓練階段，這些方法對於MLLMs的效果較差。在本研究中，我們引入了一個針對MLLMs設計的多模式數據污染檢測框架MM-Detect。我們的實驗結果表明，MM-Detect對不同程度的污染具有敏感性，並且能夠突顯由於多模式基準測試的訓練集泄漏而帶來的顯著性能改進。此外，我們還探討了污染可能源於MLLMs使用的LLMs的預訓練階段以及MLLMs的微調階段，從而提供了有關污染可能引入的階段的新見解。

English

The rapid progression of multimodal large language models (MLLMs) has demonstrated superior performance on various multimodal benchmarks. However, the issue of data contamination during training creates challenges in performance evaluation and comparison. While numerous methods exist for detecting dataset contamination in large language models (LLMs), they are less effective for MLLMs due to their various modalities and multiple training phases. In this study, we introduce a multimodal data contamination detection framework, MM-Detect, designed for MLLMs. Our experimental results indicate that MM-Detect is sensitive to varying degrees of contamination and can highlight significant performance improvements due to leakage of the training set of multimodal benchmarks. Furthermore, We also explore the possibility of contamination originating from the pre-training phase of LLMs used by MLLMs and the fine-tuning phase of MLLMs, offering new insights into the stages at which contamination may be introduced.