arXiv:2511.00056v1 Announce Type: new Abstract: The substantial memory demands of pre-training and fine-tuning large language models (LLMs) require memory-efficient optimization algorithms. One promising approach is layer-wise optimization, which treats each transformer block as a single layer and optimizes it sequentially,…
arXiv:2510.09338v2 Announce Type: replace-cross Abstract: We present a novel framework for training large language models with continuously adjustable internal representations that span the full spectrum from localist (interpretable, rule-based) to distributed (generalizable, efficient) encodings. The key innovation is a locality…
arXiv:2511.00059v1 Announce Type: new Abstract: OthelloGPT, a transformer trained to predict valid moves in Othello, provides an ideal testbed for interpretability research. The model is complex enough to exhibit rich computational patterns, yet grounded in rule-based game logic that enables…
arXiv:2511.01411v1 Announce Type: cross Abstract: Faithful yet compact explanations for vision models remain a challenge, as commonly used dense perturbation masks are often fragmented and overfitted, needing careful post-processing. Here, we present a training-free explanation method that replaces dense masks…
arXiv:2511.00064v1 Announce Type: new Abstract: Clustering algorithms often rely on restrictive assumptions: K-Means and Gaussian Mixtures presuppose convex, Gaussian-like clusters, while DBSCAN and HDBSCAN capture non-convexity but can be highly sensitive. I introduce EVINGCA (Evolving Variance-Informed Nonparametric Graph Construction Algorithm),…
arXiv:2511.01724v1 Announce Type: cross Abstract: Deep learning models are notoriously vulnerable to imperceptible perturbations. Most existing research centers on adversarial robustness (AR), which evaluates models under worst-case scenarios by examining the existence of deterministic adversarial examples (AEs). In contrast, probabilistic…
The new approach maps aircraft sections most vulnerable to lightning, including on planes with experimental designs.
arXiv:2410.11067v3 Announce Type: replace-cross Abstract: Given an intractable target density $p$, variational inference (VI) attempts to find the best approximation $q$ from a tractable family $Q$. This is typically done by minimizing the exclusive Kullback-Leibler divergence, $text{KL}(q||p)$. In practice, $Q$…
arXiv:2505.13136v2 Announce Type: replace-cross Abstract: Encoders remain essential for efficient German NLP and NLU scenarios despite the rise of decoder-only LLMs. This work studies two routes to high-quality German encoders under identical data and training constraints: 1) training from scratch…
arXiv:2507.22959v2 Announce Type: replace Abstract: The field of scientific machine learning, which originally utilized multilayer perceptrons (MLPs), is increasingly adopting Kolmogorov-Arnold Networks (KANs) for data encoding. This shift is driven by the limitations of MLPs, including poor interpretability, fixed activation…
