arXiv:2607. 25387v2 Announce Type: replace Abstract: Learned restriction maps in sheaf graph neural networks are often treated as proof that the model has discovered useful edge geometry.
arXiv:2602. 03846v2 Announce Type: replace-cross Abstract: We develop a continual learning method for pretrained models that \emph{requires no access to old-task data}, addressing a practical barrier in foundation model adaptation where pretraining distributions are often unavailable.
arXiv:2608. 01130v1 Announce Type: new Abstract: A broad range of models face the mismatch where they are updated through trajectory losses but are evaluated by downstream task reward.
arXiv:2606. 30852v1 Announce Type: new Abstract: Reasoning models spend different amounts of useful computation across instances, but it remains unclear when a learned stopping rule improves over simple confidence or convergence thresholds.
arXiv:2608. 14465v1 Announce Type: cross Abstract: A frozen language model on reasoning tasks has two coupled weaknesses: it under-uses evidence its own residual stream already encodes, and it fails to detect when the input is insufficient to answer, so it confabulates.
arXiv:2607. 25063v1 Announce Type: new Abstract: Developers judge a model checkpoint by how it behaves.
arXiv:2607. 16568v1 Announce Type: new Abstract: Function-preserving network growth techniques such as Net2Net and progressive stacking expand a model's capacity without destroying its learned function, but existing formulations either tolerate numerical perturbations or require a full rebuild of the training program.
arXiv:2606. 03808v1 Announce Type: cross Abstract: We propose PURGE, a machine unlearning algorithm built on a simple but an under-exploited observation: continual learning (CL) and machine unlearning (MU) which are fundamentally dual problems.
Safety alignment in large language models can be fragile under fine-tuning, as even benign task adaptation may increase harmful compliance. Existing defenses mainly follow two directions: they either intervene during or after fine-tuning through retraining or weight modification, which can be costly and may hurt task performance, or they use model-agnostic safety classifiers, which may miss failures specific to a given fine-tuned checkpoint.
arXiv:2604. 21395v3 Announce Type: replace-cross Abstract: Ordinary supervised training minimises the task loss and then stops.
arXiv:2607. 16821v1 Announce Type: cross Abstract: Task arithmetic, sequential fine-tuning, activation steering, and first-order random search all operate through relatively small perturbations around an already trained checkpoint, and they rely on different local approximations: individual perturbations should be first-order predictable, task updates should compose with controlled interference, useful tangent structure should be stable and possible to estimate, and weight edits should have counterparts in representation space.
arXiv:2602. 09689v2 Announce Type: replace Abstract: Fine-tuning large pre-trained models on a target distribution often improves in-distribution (ID) accuracy, but at the cost of out-of-distribution (OOD) robustness as representations specialize to the fine-tuning data.
arXiv:2608. 14717v1 Announce Type: cross Abstract: A query-relation deletion can improve the edited slot while reducing the utility of the prediction set that contains it.