arXiv Machine Learning June 18, 2026 By Joshua Nunley

Attention as Frustrated Synchronization

Read the original on arXiv Machine Learning →

arXiv:2606. 18694v1 Announce Type: new Abstract: A network of oscillators that synchronizes perfectly computes nothing further, so an attention architecture built from synchronization must locate its computation in structured departures from agreement.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

llms

arXiv Machine Learning

Jun 11

Kuramoto Attention: Synchronizing Self-Attention on the Torus

arXiv:2606. 11585v1 Announce Type: new Abstract: We introduce Kuramoto attention, a self-attention layer in which each hidden coordinate is an angle.

By Joshua Nunley

llms

Hugging Face Trending Papers

Jun 11

Circuit Synchronization Precedes Generalization: Causal Evidence from Fourier Structure in Grokking Transformers

Grokking -- where a transformer on modular arithmetic suddenly transitions from near-chance to near-perfect validation accuracy -- is attributed to a Fourier circuit, but its timing, causal structure, and controllability remain poorly understood. We introduce the Frequency Synchronization Degree (FSD), a normalised, permutation-tested metric for Fourier circuit synchronisation requiring no prior circuit knowledge.

llms

arXiv Machine Learning

Jun 11

Attention by Synchronization in Coupled Oscillator Networks

arXiv:2606. 12059v1 Announce Type: new Abstract: We address transformer attention on energy-constrained physical substrates.

By Fabio Pasqualetti, Taosha Guo

llms

arXiv Machine Learning

Jun 24

Harmonic: Hierarchical State Space Models for Efficient Long-Context Language Modeling

arXiv:2606. 24650v1 Announce Type: cross Abstract: We present Harmonic, a hierarchical state space model (SSM) for language modeling.

By Petr Nyoma

llms benchmarks

arXiv Machine Learning

Aug 12

Off-Axis, On Purpose: Where a Transformer Computes Concepts and Why it Does So

arXiv:2608. 10251v1 Announce Type: cross Abstract: A transformer's answer lives on one axis: the direction its unembedding reads.

By Mark Oskin

llms benchmarks

arXiv AI

Jul 8

Transformers converge to invariant algorithmic cores

arXiv:2602. 22600v2 Announce Type: replace-cross Abstract: Training selects for behavior, not circuitry: many weight configurations can implement the same function.

By Joshua S. Schiffman

llms

One email a morning, machine-written

Related stories

Kuramoto Attention: Synchronizing Self-Attention on the Torus

Circuit Synchronization Precedes Generalization: Causal Evidence from Fourier Structure in Grokking Transformers

Attention by Synchronization in Coupled Oscillator Networks

Harmonic: Hierarchical State Space Models for Efficient Long-Context Language Modeling

Off-Axis, On Purpose: Where a Transformer Computes Concepts and Why it Does So

Transformers converge to invariant algorithmic cores