arXiv:2511.20798v1 Announce Type: new Abstract: Recent advances in mechanistic interpretability have revealed that large language models (LLMs) develop internal representations corresponding not only to concrete entities but also distinct, human-understandable abstract concepts and behaviour. Moreover, these hidden features can be…
arXiv:2406.19622v2 Announce Type: replace Abstract: As deep neural networks (DNNs) are increasingly deployed in sensitive applications, ensuring their security and robustness has become critical. A major threat to DNNs arises from adversarial attacks, where small input perturbations can lead to…
arXiv:2511.20811v1 Announce Type: new Abstract: We develop a data-driven approach for runtime safety monitoring in flight testing, where pilots perform maneuvers on aircraft with uncertain parameters. Because safety violations can arise unexpectedly as a result of these uncertainties, pilots need…
arXiv:2505.18636v2 Announce Type: replace Abstract: The go-to strategy to apply deep networks in settings where uncertainty informs decisions–ensembling multiple training runs with random initializations–is ill-suited for the extremely large-scale models and practical fine-tuning workflows of today. We introduce a new…
arXiv:2511.20826v1 Announce Type: new Abstract: Untrained large neural networks, just after random initialization, tend to favour a small subset of classes, assigning high predicted probabilities to these few classes and approximately zero probability to all others. This bias, termed Initial…
arXiv:2511.20830v1 Announce Type: new Abstract: The solar wind, a continuous outflow of charged particles from the Sun’s corona, shapes the heliosphere and impacts space systems near Earth. Accurate prediction of features such as high-speed streams and coronal mass ejections is…
arXiv:2511.10234v2 Announce Type: replace Abstract: While promising, graph reasoners based on Large Language Models (LLMs) lack built-in invariance to symmetries in graph representations. Operating on sequential graph serializations, LLMs can produce different outputs under node reindexing, edge reordering, or formatting…
arXiv:2511.20839v1 Announce Type: new Abstract: We present Primal, a deterministic feature mapping framework that harnesses the number-theoretic independence of prime square roots to construct robust, tunable vector representations. Diverging from standard stochastic projections (e.g., Random Fourier Features), our method exploits…
arXiv:2511.20099v2 Announce Type: replace Abstract: Large language models (LLMs) have shown promising capabilities in hardware description language (HDL) generation. However, existing approaches often rely on free-form natural language descriptions that are often ambiguous, redundant, and unstructured, which poses significant challenges…
arXiv:2511.20844v1 Announce Type: new Abstract: Training deep networks with noisy labels leads to poor generalization and degraded accuracy due to overfitting to label noise. Existing approaches for learning with noisy labels often rely on the availability of a clean subset…
