The scripts below execute standard baseline unlearning experiments on the TOFU and MUSE datasets, evaluated using their corresponding benchmarks.
bash scripts/tofu_unlearn.sh
bash scripts/muse_unlearn.sh
For all the experiments below, we used the following setup
Category | Details |
---|---|
Hardware | 2 × L40s GPUs (48GB each) |
Distributed Computing | DeepSpeed ZeRO Stage 3 (Accelerate) |
Hyperparameters | Learning Rate (lr) = 1e-5 α = 1, γ = 1, β = 0.1 (where applicable) Number of Epochs = 10 Optimizer: paged_adamw_32bit |
Note:
- Results may vary even with the same effective hyperparameters when trained with modifications to the distributed training setup, including when training on a single GPU. For example: methods such as SimNPO, can be significantly improved with careful tuning. Please use these numbers only for reproducibility purposes.
- NPO in MUSE: for NPO, the MUSE implementation is inconsistent with the original paper as discussed here. This inconsistency is carried over into implementations like SimNPO. Here, we use the original NPO implementation with the same loss function expression across datasets.
Method | forget01 | forget05 | forget10 | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
forget_quality | model_utility | forget_truth_ratio | forget_quality | model_utility | forget_truth_ratio | forget_quality | model_utility | forget_truth_ratio | |||||||||||
Finetuned | 1.27e-03 | 0.63 | 0.53 | 1.33e-13 | 0.63 | 0.51 | 4.35e-25 | 0.63 | 0.52 | ||||||||||
Retain | 0.0 | 0.63 | 0.68 | 0 | 0.63 | 0.67 | 0.0 | 0.61 | 0.68 | ||||||||||
GradAscent | 1.88e-04 | 0.55 | 0.36 | 1.94e-119 | 0.00e+00 | 8.82e-96 | 1.06e-239 | 0.00e+00 | 2.21e-32 | ||||||||||
GradDiff | 3.02e-03 | 0.57 | 0.41 | 1.94e-119 | 0.56 | 4.14e-95 | 1.80e-229 | 0.58 | 1.46e-07 | ||||||||||
IdkDPO | 0.1 | 0.56 | 0.67 | 4.02e-06 | 0.04 | 0.67 | 5.42e-13 | 0.04 | 0.64 | ||||||||||
NPO | 0.4 | 0.58 | 0.65 | 0.09 | 0.53 | 0.71 | 0.42 | 0.54 | 0.73 | ||||||||||
SimNPO | 1.27e-03 | 0.58 | 0.41 | 1.06e-106 | 0.6 | 3.94e-05 | 1.47e-198 | 0.6 | 3.17e-04 |
Method | forget01 | forget05 | forget10 | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
forget_quality | model_utility | forget_truth_ratio | forget_quality | model_utility | forget_truth_ratio | forget_quality | model_utility | forget_truth_ratio | |||||||||||
Finetuned | 0.01 | 0.60 | 0.47 | 2.96e-13 | 0.6 | 0.47 | 8.08e-22 | 0.6 | 0.48 | ||||||||||
Retain | 0 | 0.60 | 0.65 | 0 | 0.6 | 0.63 | 0 | 0.59 | 0.63 | ||||||||||
GradAscent | 0.27 | 0.33 | 0.59 | 1.94e-119 | 0 | 2.52e-23 | 1.06e-239 | 0 | 2.25e-18 | ||||||||||
GradDiff | 0.77 | 0.43 | 0.57 | 1.94e-119 | 0.53 | 3.87e-34 | 1.06e-239 | 0.49 | 3.53e-27 | ||||||||||
IdkDPO | 0.01 | 0.51 | 0.60 | 1.12e-05 | 0.07 | 0.62 | 4.64e-12 | 0.23 | 0.6 | ||||||||||
NPO | 0.92 | 0.56 | 0.66 | 0.14 | 0.45 | 0.7 | 0.02 | 0.46 | 0.7 | ||||||||||
SimNPO | 0.58 | 0.46 | 0.55 | 5.01e-100 | 0.58 | 4.19e-03 | 2.47e-203 | 0.54 | 1.07e-05 |
Method | News | Books | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
forget_knowmem_ROUGE | forget_verbmem_ROUGE | privleak | retain_knowmem_ROUGE | forget_knowmem_ROUGE | forget_verbmem_ROUGE | privleak | retain_knowmem_ROUGE | ||||||||||||
Finetuned | 0.64 | 0.58 | -99.81 | 0.55 | 0.47 | 1.0 | -57.26 | 0.69 | |||||||||||
Retain | 0.33 | 0.21 | -4.54 | 0.56 | 0.3 | 0.14 | 7.96 | 0.69 | |||||||||||
GradAscent | 0 | 0 | 52.11 | 0 | 0 | 0 | -0.67 | 0 | |||||||||||
GradDiff | 0.41 | 8.92e-03 | 93.23 | 0.37 | 0.18 | 0.16 | -37.79 | 0.3 | |||||||||||
NPO | 0.56 | 0.35 | -86.00 | 0.51 | 0.32 | 0.84 | -54.24 | 0.55 | |||||||||||
SimNPO | 0.54 | 0.36 | -86.11 | 0.51 | 0.32 | 0.84 | -54.26 | 0.54 |