Renormalization Group

Follow this topic Ask about this topic

Latest papers

11 papers3 papers in the last four weeks, with none the four weeks before. 0.0% of all new papers.

Information-Theoretic Reliability--Safety Bounds and Finite-Alignment Dynamics in AI Knowledge Distillation
Oct 28, 2026Miao-Miao Yi, Zhen-Yu Cai, Y -M. Du +1Renormalization Group Trustworthy AI
Graduate School of China Academy of Engineering Physics, Beijing, 100193, China
Scaling and Renormalization in Neural Networks: An Effective Renormalization-Group Description of Grokking
Oct 26, 2026Min Chen, Junqi Wang, Zeguan Wu +2Renormalization Group Grokking
Department of Computer Science, University of Pittsburgh, Pittsburgh, PA, USA
Symbolic Density Estimators for Unnormalized Distributions
Oct 7, 2026Vikas Kanaujia, Riyansha Singh, Shashank Sharma +1Renormalization Group Density Estimation
Department of Electrical Engineering IIT Kanpur · Department of Computer Science and Engineering IIT Kanpur · Department of Physics IIT Kanpur +1
Neural Renormalization Group Flow for Percolation
Aug 27, 2026Anaclara Alvez, Luca Camagna, Sergio Chibbaro +4Renormalization Group Phase Transitions
LISN, AO team, Bˆat 660 Universit´e Paris-Saclay, Orsay Cedex 91405 · Inria Saclay - Tau team, Bˆat 660 Universit´e Paris-Saclay, Orsay Cedex 91405
Renormalization Group Flow Matching for Scalable Local Generative Modeling
Aug 24, 2026Kanta Masuki, Yuto AshidaRenormalization Group Generative Modeling
Department of Physics, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan · Institute for Physics of Intelligence, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
Gravitational Duals from Equations of State II: Large Hierarchies and False Vacua
Jun 29, 2026Raul Jimenez, David Mateos, Pavlos Protopapas +3Renormalization Group High-Energy Physics
Institut de Ciències del Cosmos (ICC), Universitat de Barcelona, Martí i Franquès 1, ES-08028, Barcelona, Spain. · Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig Lluís Companys 23, ES-08010, Barcelona, Spain. · Institute for Applied Computational Science, Harvard University, Cambridge, MA.
Rank Collapse, Fixed Points, and the Renormalization Group Structure of MLP Residual Networks
Jun 9, 2026Parviz Haggi-Mani, Irina RishMultilayer Perceptrons Renormalization Group
Universit´e de Montr´eal · Mila - Quebec AI Institute
Interpreting FCDNNs via RG on Exponential Family
May 29, 2026Fuzhou Gong, Zigeng XiaRenormalization Group Neural Network Interpretability
Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China. · School of Mathematics, Northwest University, Xi’an 710127, China.
Lecture Notes on Statistical Physics and Neural Networks
May 7, 2026Olaf HohmPhase Transitions Renormalization Group
Institute for Physics, Humboldt University Berlin, Zum Großen Windkanal 2, D-12489 Berlin, Germany
A renormalization-group inspired lattice-based framework for piecewise generalized linear models
May 6, 2026Joshua C. ChangInterpretable ML Renormalization Group
NIH Clinical Center, Rehabilitation Medicine · Mederrata Research Inc · Sound Prediction Inc
Machine learning for four-dimensional SU(3) lattice gauge theories
Apr 14, 2026Urs WengerLattice Field Theory High-Energy Physics
Institute for Theoretical Physics Albert Einstein Center for Fundamental Physics University of Bern CH-3012 Bern Switzerland

Oct 28, 2026J/K move · Enter open · S save

Information-Theoretic Reliability--Safety Bounds and Finite-Alignment Dynamics in AI Knowledge Distillation

Miao-Miao Yi, Zhen-Yu Cai, Y -M. Du, C. P. Sun

Graduate School of China Academy of Engineering Physics, Beijing, 100193, China

Renormalization Group Trustworthy AI Knowledge Distillation

Open paper AskPDF arXiv

The reliability and safety of artificial intelligence (AI), together with their transfer and evolution through knowledge distillation, have attracted growing attention. We establish an information-theoretic bound on the reliability--safety trade-off and characterize the dynamics of this trade-off under knowledge distillation. We first prove a universal trade-off inequality that is independent of the specific model form. Its boundary is determined solely by a single parameter Kh, which quantifies the hidden layer's ability to distinguish benign from hazardous inputs. We then develop a canonical Boltzmann-machine model of AI response behavior and derive two exact relations between reliability and safety. The trade-off curve is determined by the hazard discrimination capacity K, while the overall answer tendency controls the operating point along the curve. These relations agree well with existing experimental data. In knowledge distillation, we find that teacher responses induce a renormalization of the student model's trade-off parameters. We further use a renormalization-group approach to study multigenerational distillation, deriving exact recursion relations and the associated renormalization-group-like flows for two representative direct intergenerational distillation protocols at arbitrary teacher--student alignment strength. These flows reveal fixed-point structures, long-term dynamics, and critical behavior, yielding testable predictions for the critical transition between capability loss and stable transmission.