LLM tabular generators leak memorized numeric strings, allowing a no-box attack to achieve near-perfect membership inference on some state-of-the-art models.
arXiv preprint arXiv:2406.05216 , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.LG 3representative citing papers
TabGRAA applies group-relative advantage alignment in an iterative reward-guided post-training loop to improve tabular language model generators on fidelity, utility, and privacy trade-offs across five benchmarks.
Thresholding and downsampling effectively mitigate class imbalance in PFNs for tabular classification due to their calibration and limited-data strengths.
citing papers explorer
-
When Tables Leak: Attacking String Memorization in LLM-Based Tabular Data Generation
LLM tabular generators leak memorized numeric strings, allowing a no-box attack to achieve near-perfect membership inference on some state-of-the-art models.
-
Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Training
TabGRAA applies group-relative advantage alignment in an iterative reward-guided post-training loop to improve tabular language model generators on fidelity, utility, and privacy trade-offs across five benchmarks.
-
Correcting Class Imbalance in Prior-Data Fitted Networks for Tabular Classification
Thresholding and downsampling effectively mitigate class imbalance in PFNs for tabular classification due to their calibration and limited-data strengths.