arxiv1h agobullish

H-Sets: Hessian-Guided Discovery of Set-Level Feature Interactions in Image Classifiers

arXiv:2604.22045v1 Announce Type: cross Abstract: Feature attribution methods explain the predictions of deep neural networks by assigning importance scores to individual input features. However, most existing methods focus solely on marginal effects, overlooking feature interactions, where groups o

VGREDE5 models · +2 #computer-vision #interpretability

arxiv4d ago

LayerTracer: A Joint Task-Particle and Vulnerable-Layer Analysis framework for Arbitrary Large Language Model Architectures

arXiv:2604.20556v1 Announce Type: cross Abstract: Currently, Large Language Models (LLMs) feature a diversified architectural landscape, including traditional Transformer, GateDeltaNet, and Mamba. However, the evolutionary laws of hierarchical representations, task knowledge formation positions, and

TRGAMA3 models #large-language-models #architecture #interpretability Read on arxiv →

arxivApr 18

Structural interpretability in SVMs with truncated orthogonal polynomial kernels

arXiv:2604.15285v1 Announce Type: cross Abstract: We study post-training interpretability for Support Vector Machines (SVMs) built from truncated orthogonal polynomial kernels. Since the associated reproducing kernel Hilbert space is finite-dimensional and admits an explicit tensor-product orthonorm

SU1 model #machine-learning #interpretability #kernel-methods Read on arxiv →

arxivApr 14

Principles Do Not Apply Themselves: A Hermeneutic Perspective on AI Alignment

arXiv:2604.10673v1 Announce Type: new Abstract: AI alignment is often framed as the task of ensuring that an AI system follows a set of stated principles or human preferences, but general principles rarely determine their own application in concrete cases. When principles conflict, when they are too

#alignment #interpretability #evaluation Read on arxiv →

arxivApr 9

Continuous Interpretive Steering for Scalar Diversity

arXiv:2604.07006v1 Announce Type: new Abstract: Pragmatic inference is inherently graded. Different lexical items give rise to pragmatic enrichment to different degrees. Scalar implicature exemplifies this property through scalar diversity, where implicature strength varies across scalar items. Howe

LA1 model #pragmatic-inference #language-models #interpretability Read on arxiv →