跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2605.10931 · v3

Quantifying Concentration Phenomena of Mean-Field Transformers in the Low-Temperature Regime

首次提交 2026-05-12 01:58
版本更新 2026-11-17 23:54
雷达观测 2026-11-18 16:38
arXiv ↗PDF ↗

摘要

Transformers with self-attention modules as their core components have become an integral architecture in modern large language and foundation models. In this paper, we study the evolution of tokens in deep encoder-only transformers at inference time which is described in the large-token limit by a mean-field continuity equation. Leveraging ideas from the convergence analysis of interacting multi-particle systems, with particles corresponding to tokens, we prove that the token distribution rapidly concentrates onto the push-forward of the initial distribution under a projection map induced by the key, query, and value matrices, and remains metastable for moderate times. Specifically, we show that the Wasserstein distance of the two distributions scales like log⁡(β+1)/βexp⁡(Ct)+exp⁡(−ct)\sqrt{{\log(β+1)}/β}\exp(Ct)+\exp(-ct) in terms of the temperature parameter β−1→0β^{-1}\to 0 and inference time t≥0t\geq 0. For the proof, we establish Lyapunov-type estimates for the zero-temperature equation, identify its limit as t→∞t\to\infty, and employ a stability estimate in Wasserstein space together with a quantitative Laplace principle to couple the two equations. Our result implies that for time scales of order log⁡β\logβ, which correspond to the early-phase behavior most relevant for practical finite-depth models, the token distribution concentrates at the identified limiting distribution. Numerical experiments confirm this and, beyond that, complement our theory by showing that for finite ββ and large tt the dynamics enter a different terminal phase, dominated by the spectrum of the value matrix.

作者备注:Accepted at the NeurIPS 2026 Conference: https://neurips.cc/virtual/2026/loc/paris/poster/149032 Funded by the European Union's HORIZON-MSCA-2021-DN-01-01, through ModConFlex ("Modelling and control of flexible structures interacting with fluids"; No. 101073558; DOI: 10.3030/101073558). Code: https://github.com/albertalcalde/mean-field-transformer-concentration, code DOI: 10.5281/zenodo.23726583

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v3,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v3 · 2026-11-17 23:54 · 最新观测版本

Quantifying Concentration Phenomena of Mean-Field Transformers in the Low-Temperature Regime

Transformers with self-attention modules as their core components have become an integral architecture in modern large language and foundation models. In this paper, we study the evolution of tokens in deep encoder-only transformers at inference time which is described in the large-token limit by a mean-field continuity equation. Leveraging ideas from the convergence analysis of interacting multi-particle systems, with particles corresponding to tokens, we prove that the token distribution rapidly concentrates onto the push-forward of the initial distribution under a projection map induced by the key, query, and value matrices, and remains metastable for moderate times. Specifically, we show that the Wasserstein distance of the two distributions scales like log⁡(β+1)/βexp⁡(Ct)+exp⁡(−ct)\sqrt{{\log(β+1)}/β}\exp(Ct)+\exp(-ct) in terms of the temperature parameter β−1→0β^{-1}\to 0 and inference time t≥0t\geq 0. For the proof, we establish Lyapunov-type estimates for the zero-temperature equation, identify its limit as t→∞t\to\infty, and employ a stability estimate in Wasserstein space together with a quantitative Laplace principle to couple the two equations. Our result implies that for time scales of order log⁡β\logβ, which correspond to the early-phase behavior most relevant for practical finite-depth models, the token distribution concentrates at the identified limiting distribution. Numerical experiments confirm this and, beyond that, complement our theory by showing that for finite ββ and large tt the dynamics enter a different terminal phase, dominated by the spectrum of the value matrix.

查看此版本解读 →查看此版本原文 ↗