notesum.ai
Published at November 10vTune: Verifiable Fine-Tuning for LLMs Through Backdooring
cs.LG
cs.AI
cs.CY
Released Date: November 10, 2024
Authors: Eva Zhang1, Arka Pal1, Akilesh Potti1, Micah Goldblum2
Aff.: 1Ritual; 2Columbia University

| Dataset | p-values | Likelihood of signature without fine-tuning | ||
| RecipeNLG | 10000 | 50 | 1.85e-45 | 0.00 |
| MathInstruct | 10000 | 50 | 2.03e-42 | 0.00 |
| ShareGPT | 15000 | 470 | 6.81e-46 | 0.00 |
| SQuAD | 87400 | 437 | 6.05e-39 | 0.00 |
| XLSum | 7200 | 36 | 9.85e-34 | 0.00 |
| MedQA | 10200 | 51 | 1.11e-40 | 0.00 |
| CodeFeedback | 10050 | 50 | 1.50e-41 | 0.00 |