Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Min-K%++: Improved Baseline for Detecting Pre-Training Data from Large Language Models

About

The problem of pre-training data detection for large language models (LLMs) has received growing attention due to its implications in critical issues like copyright violation and test data contamination. Despite improved performance, existing methods (including the state-of-the-art, Min-K%) are mostly developed upon simple heuristics and lack solid, reasonable foundations. In this work, we propose a novel and theoretically motivated methodology for pre-training data detection, named Min-K%++. Specifically, we present a key insight that training samples tend to be local maxima of the modeled distribution along each input dimension through maximum likelihood training, which in turn allow us to insightfully translate the problem into identification of local maxima. Then, we design our method accordingly that works under the discrete distribution modeled by LLMs, whose core idea is to determine whether the input forms a mode or has relatively high probability under the conditional categorical distribution. Empirically, the proposed method achieves new SOTA performance across multiple settings. On the WikiMIA benchmark, Min-K%++ outperforms the runner-up by 6.2% to 10.5% in detection AUROC averaged over five models. On the more challenging MIMIR benchmark, it consistently improves upon reference-free methods while performing on par with reference-based method that requires an extra reference model.

Jingyang Zhang, Jingwei Sun, Eric Yeats, Yang Ouyang, Martin Kuo, Jianyi Zhang, Hao Frank Yang, Hai Li• 2024

Related benchmarks

TaskDatasetResultRank
Membership InferenceWikiMIA 32 tokens 1.0
ROC AUC84.8
66
Suffix RankingExtraction Challenge Dataset
MP (%)47.6
66
Membership Inference AttackAsclepius (fine-tuned)
TPR@FPR=0.011.82
58
Membership Inference AttackWikipedia
AUC0.642
52
Membership Inference AttackAG News (test)
AUC0.604
43
Membership Inference AttackXSum (test)
AUC0.73
43
Membership Inference AttackHealthcareMagic
AUC79.8
36
Membership Inference AttackCNN/DM
AUC0.861
36
Membership Inference AttackMedInstruct
AUC81
36
Membership Inference AttackWikiMIA-25
AUC0.727
33
Showing 10 of 51 rows

Other info

Follow for update