{"ID":3006076,"CreatedAt":"2026-06-03T03:09:48.883664427Z","UpdatedAt":"2026-06-04T19:14:31.964469513Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.02860","arxiv_id":"2606.02860","title":"Forgetting is Not Erasure: Recovering Latent Knowledge via Transport Keys","abstract":"Catastrophic forgetting is often framed as a representational problem: after sequential training, a model appears to lose the features that supported performance on earlier tasks. We challenge the stronger form of this view. Across controlled continual-learning settings, we find that a significant portion of apparent forgetting can be attributed to interface drift between internal stages rather than permanent erasure of task-relevant computation. We study this phenomenon through a stitched evaluation protocol that combines early computation from a post-update network with late computation from its predecessor, optionally mediated by a compact, task-specific transport key. We describe transport keys at a systems level as compact interface-alignment operators estimated from a small set of paired anchor activations and evaluated through model stitching. On split CIFAR-100 with a ResNet-style network, transport keys recover most of the original Task A performance after sequential training on Task B. On a compact vision transformer, we observe a similar recovery pattern. These results suggest that continual learning may require better mechanisms for indexing and re-accessing latent computations, not only methods that prevent weight change.","short_abstract":"Catastrophic forgetting is often framed as a representational problem: after sequential training, a model appears to lose the features that supported performance on earlier tasks. We challenge the stronger form of this view. Across controlled continual-learning settings, we find that a significant portion of apparent f...","url_abs":"https://arxiv.org/abs/2606.02860","url_pdf":"https://arxiv.org/pdf/2606.02860v1","authors":"[\"Archie Chaudhury\"]","published":"2026-06-01T20:22:03Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Vision Transformer\",\"Transformer\"]","has_code":false}