🏢 Heidelberg University
Truth is Universal: Robust Detection of Lies in LLMs
·4200 words·20 mins·
loading
·
loading
Natural Language Processing
Large Language Models
🏢 Heidelberg University
LLM lie detectors fail to generalize; this paper presents a robust method achieving 94% accuracy by identifying a universal two-dimensional truth subspace, separating true/false statements across vari…
Lorentz-Equivariant Geometric Algebra Transformers for High-Energy Physics
·1502 words·8 mins·
loading
·
loading
Machine Learning
Deep Learning
🏢 Heidelberg University
Lorentz Geometric Algebra Transformer (L-GATr): A novel, scalable architecture for high-energy physics, achieving high-precision, data-efficient learning and outperforming existing methods on regressi…
Learning Distributions on Manifolds with Free-Form Flows
·2354 words·12 mins·
loading
·
loading
Machine Learning
Generative Models
🏢 Heidelberg University
Manifold Free-Form Flows (M-FFF) achieves fast and accurate generative modeling on Riemannian manifolds using a single function evaluation, outperforming prior methods.