Published Unlearning Numbers Move Per Checkpoint, and Not Because the Removed Data Survives: An Audit of 263 Released Batch-Normalized Checkpoints

The audit examines 263 batch‑normalized checkpoints released by arXiv, finding that refitting models on retained data at identical weights shifts 47 of 221 checkpoints beyond the spread indicated by their own release seeds. This movement is attributed to checkpoint properties rather than the survival of removed data, as swapping removed records for kept ones barely changes the published state. The study concludes that releases should specify the fitting convention used, especially for batch‑normalized vision models.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.