Dongyoon Han, Junsuk Choe, Hyo Kim
We lifted 6 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| naver-ai/negmerge | canonical | 3 of 6 |
| Function | Status | Where it lives |
|---|---|---|
| cosine_lr | Ran | naver-ai/negmerge/CLIP_MU/src/utils.py pointer only (licence: NONE) · get_code("2d5d7a12d8de02b7") |
| custom_scaled_dot_product_attention | Ran | naver-ai/negmerge/CLIP_MU/src/modeling.py pointer only (licence: NOASSERTION) · get_code("7d9de96609777711") |
| get_module_by_name | Ran | naver-ai/negmerge/CLIP_MU/src/modeling.py pointer only (licence: NOASSERTION) · get_code("23886ccd64a06f48") |
| accuracy | Not yet run | naver-ai/negmerge/CLIP_MU/src/utils.py pointer only (licence: NONE) · get_code("aafcff3ad1ccef6e") |
| distribute_loader | Not yet run | naver-ai/negmerge/CLIP_MU/src/distributed.py pointer only (licence: NOASSERTION) · get_code("57acc0bee3da475a") |
| torch_load_old | Not yet run | naver-ai/negmerge/CLIP_MU/src/utils.py pointer only (licence: NONE) · get_code("7d051a7adefedb6c") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Machine unlearning aims to selectively remove specific knowledge from a trained model. Existing approaches, such as Task Arithmetic, finetune the model on the forget set to create a task vector (i.e., a direction in weight space) for subtraction from the original model's weight. However, their effectiveness is highly sensitive to hyperparameter selection, requiring extensive validation to identify the optimal vector from many fine-tuned candidates. In this paper, we propose a novel method that utilizes all fine-tuned models trained with varying hyperparameters instead of a single selection. Specifically, we aggregate the computed task vectors by retaining only the elements with consistent shared signs. The merged task vector is then negated to induce unlearning on the original model. Evaluations on zero-shot and standard image recognition tasks across twelve datasets and four backbone architectures show that our approach outperforms state-ofthe-art methods while requiring similar or fewer computational resources. Code is available at https://github.com/naver-ai/negmerge.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2410.05583")
get_code_for_paper("2410.05583")
have("2410.05583")
Connect an agent — have() is free.