# Phase transitions in Doi-Onsager, Noisy Transformer, and other multimodal models

arXiv: https://arxiv.org/abs/2604.16288 · HTML: https://mlsanity.com/paper/2604.16288/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 27 references that are in the map.
- First posted on arXiv 17 Apr 2026.

## Builds on (most cited first, 4 of 4)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [A mathematical perspective on Transformers](https://mlsanity.com/paper/2312.10794/) (2023)
- [On the Structure of Stationary Solutions to McKean-Vlasov Equations with Applications to Noisy Transformers](https://mlsanity.com/paper/2510.20094/) (2025)
- [Noisy Hegselmann-Krause Systems: Phase Transition and the 2R-Conjecture](https://mlsanity.com/paper/1511.02975/) (2015)

## Built on by (most cited first, 2 of 2)

- [On the Diverse Dynamical Behaviors Arising in Deep Linear Transformers](https://mlsanity.com/paper/2607.18584/) (2026)
- [Phase transitions for the noisy transformer model in arbitrary dimension](https://mlsanity.com/paper/2606.05140/) (2026)

## Abstract (arXiv)

We study phase transitions for repulsive-attractive mean-field free energies on the circle. For a frac1n+1-periodic interaction whose Fourier coefficients satisfy a certain decay condition, we prove that the critical coupling strength Kc coincides with the linear stability threshold K_# of the uniform distribution and that the phase transition is continuous, in the sense that the uniform distribution is the unique global minimizer at criticality. The proof is based on a sharp coercivity estimate for the free energy obtained from the constrained Lebedev--Milin inequality. We apply this result to three motivating models for which the exact value of the phase transition and its (dis)continuity in terms of the model parameters was not fully known. For the two-dimensional Doi--Onsager model W(θ)=-|sin(2πθ)|, we prove that the phase transition is continuous at Kc=K_#=3π/4. For the noisy transformer model Wβ(θ)=(eβcos(2πθ)-1)/β, we identify the sharp threshold β_* such that Kc(β) = K_#(β) and the phase transition is continuous for β ≤ β_*, while Kc(β)<K_#(β) and the phase transition is discontinuous for β > β_*. We also obtain the corresponding sharp dichotomy for the noisy Hegselmann--Krause model WR(θ) = (R-2π|θ|)+2 .