Rewriting Memory: an Efficient Framework for Machine Unlearning in Pretrained Deep Models
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
With increasing focus on data privacy, machine unlearning-the operation of deleting specific data influence from learned models-has been subject to greater scrutiny. This paper introduces a real-world machine unlearning approach from pretrained deep models that addresses data deletion issues without full retraining. Our method deletes 93.2 % of data influence with more than 98.1 % of model accuracy retained on the rest of the data. Experiments on standard datasets such as CIFAR-10 and ImageNet have at most 60 % reduced computational cost over relearning from scratch. The method combines knowledge localization and selective memory editing, and it is policy-compatible with regulations such as GDPR. Results indicate the possibility of scalable, privacy-compatible AI systems that can learn-and forget-in an efficient and ethical way.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it