training-data-detection

Tag

Cards List
#training-data-detection

Weak Ties, Strong Signals: Efficient Training Data Detection in Diffusion LLMs via Independent Token Sampling

arXiv cs.LG ↗ · 2026-09-22 Cached

The paper proposes Independent Token Sampling (ITS), a query-efficient framework for detecting training data in diffusion large language models by selecting weakly dependent tokens to reduce approximation errors, achieving improved performance over state-of-the-art baselines.

0 favorites 0 likes
← Back to home

Submit Feedback