Cover Image

Grokking as a Falsifiable Finite-Size Transition

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Bi, Yuda, Zhang, Chenyu, Wang, Qiheng, Calhoun, Vince D
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Statistical Mechanics Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Cross-Spectral Witness for Hidden Nonequilibrium Beyond the Scalar Ceiling
by: Bi, Yuda, et al.
Published: (2026)

Conditioning on a Volatility Proxy Compresses the Apparent Timescale of Collective Market Correlation
by: Bi, Yuda, et al.
Published: (2026)

Timescale Coalescence Makes Hidden Persistent Forcing Spectrally Dark
by: Bi, Yuda, et al.
Published: (2026)

Scaling Laws are Redundancy Laws
by: Bi, Yuda, et al.
Published: (2025)

Reservoir Subspace Injection for Online ICA under Top-n Whitening
by: Xiao, Wenjun, et al.
Published: (2026)

Redundancy as a Structural Information Principle for Learning and Generalization
by: Bi, Yuda, et al.
Published: (2025)

A 1/R Law for Kurtosis Contrast in Balanced Mixtures
by: Bi, Yuda, et al.
Published: (2026)

Spectral Coherence Index: A Model-Free Metric for Protein Structural Ensemble Quality Assessment
by: Bi, Yuda, et al.
Published: (2026)

Phase Transitions in the Output Distribution of Large Language Models
by: Arnold, Julian, et al.
Published: (2024)

Trees to Flows and Back: Unifying Decision Trees and Diffusion Models
by: Ramachandran, Sai Niranjan, et al.
Published: (2026)

Spontaneous symmetry breaking and Goldstone modes for deep information propagation
by: Iqbal, Nabil, et al.
Published: (2026)

On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective
by: Li, Yuhao, et al.
Published: (2026)

Thermodynamic Irreversibility of Training Algorithms
by: Ziyin, Liu, et al.
Published: (2026)

Reinforcement Learning for Microcanonical Graph Ensemble with Assortativity Constraints
by: Choi, Hoyun, et al.
Published: (2026)

Symmetry and Generalisation in Neural Approximations of Renormalisation Transformations
by: Ashworth, Cassidy, et al.
Published: (2025)

Cross-fluctuation phase transitions reveal sampling dynamics in diffusion models
by: Ramachandran, Sai Niranjan, et al.
Published: (2025)

Stochastic Resetting Mitigates Latent Gradient Bias of SGD from Label Noise
by: Bae, Youngkyoung, et al.
Published: (2024)

Maximum diffusion reinforcement learning
by: Berrueta, Thomas A., et al.
Published: (2023)

Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics
by: Sanokowski, Sebastian, et al.
Published: (2025)

Entropy, concentration, and learning: a statistical mechanics primer
by: Balsubramani, Akshay
Published: (2024)

Temporal Memory for Resource-Constrained Agents: Continual Learning via Stochastic Compress-Add-Smooth
by: Chertkov, Michael
Published: (2026)

Tensor tree learns hidden relational structures in data to construct generative models
by: Harada, Kenji, et al.
Published: (2024)

Sampling Decisions
by: Chertkov, Michael, et al.
Published: (2025)

Mixing Artificial and Natural Intelligence: From Statistical Mechanics to AI and Back to Turbulence
by: Chertkov, Michael
Published: (2024)

Adaptive Path Integral Diffusion: AdaPID
by: Chertkov, Michael, et al.
Published: (2025)

IsingFormer: Augmenting Parallel Tempering With Learned Proposals
by: Bunaiyan, Saleh, et al.
Published: (2025)

Relaxation-assisted reverse annealing on nonnegative/binary matrix factorization
by: Haba, Renichiro, et al.
Published: (2025)

On the Separability of Information in Diffusion Models
by: Premkumar, Akhil
Published: (2025)

Generative Stochastic Optimal Transport: Guided Harmonic Path-Integral Diffusion
by: Chertkov, Michael
Published: (2025)

Pseudo Transitions in the Finite-Size Blume-Capel Model
by: Shi, Lei, et al.
Published: (2024)

Concurrence of Symmetry Breaking and Nonlocality Phase Transitions in Diffusion Models
by: Zhang, Yifan F., et al.
Published: (2026)

Estimating Global Input Relevance and Enforcing Sparse Representations with a Scalable Spectral Neural Network Approach
by: Chicchi, Lorenzo, et al.
Published: (2024)

Efficient Identification of Critical Transitions via Flow Matching: A Scalable Generative Approach for Many-Body Systems
by: Lee, Qian-Rui, et al.
Published: (2025)

Phase Transitions as the Breakdown of Statistical Indistinguishability
by: Narita, Taiyo, et al.
Published: (2026)

Breaking QAOA's Fixed Target Hamiltonian Barrier: A Fully Connected Quantum Boltzmann Machine via Bilevel Optimization
by: Liu, Jun
Published: (2026)

Analytic Bridge Diffusions for Controlled Path Generation
by: Chertkov, Michael
Published: (2026)

Reinforcement Learning for Quantum Technology
by: Bukov, Marin, et al.
Published: (2026)

A Minimal Model of Representation Collapse: Frustration, Stop-Gradient, and Dynamics
by: Yao, Louie Hong, et al.
Published: (2026)

Entropic Confinement and Mode Connectivity in Overparameterized Neural Networks
by: Di Carlo, Luca, et al.
Published: (2025)

Support Vector Machine Kernels as Quantum Propagators
by: Kuo, Nan-Hong, et al.
Published: (2025)