Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash 99 tok/s
Gemini 2.5 Pro 43 tok/s Pro
GPT-5 Medium 28 tok/s
GPT-5 High 35 tok/s Pro
GPT-4o 94 tok/s
GPT OSS 120B 476 tok/s Pro
Kimi K2 190 tok/s Pro
2000 character limit reached

Thermalization and canonical typicality in translation-invariant quantum lattice systems (1312.7420v5)

Published 28 Dec 2013 in quant-ph, cond-mat.stat-mech, math-ph, and math.MP

Abstract: It has previously been suggested that small subsystems of closed quantum systems thermalize under some assumptions; however, this has been rigorously shown so far only for systems with very weak interaction between subsystems. In this work, we give rigorous analytic results on thermalization for translation-invariant quantum lattice systems with finite-range interaction of arbitrary strength, in all cases where there is a unique equilibrium state at the corresponding temperature. We clarify the physical picture by showing that subsystems relax towards the reduction of the global Gibbs state, not the local Gibbs state, if the initial state has close to maximal population entropy and certain non-degeneracy conditions on the spectrum are satisfied. Moreover, we show that almost all pure states with support on a small energy window are locally thermal in the sense of canonical typicality. We derive our results from a statement on equivalence of ensembles generalizing earlier results by Lima, and give numerical and analytic finite-size bounds, relating the Ising model to the finite de Finetti theorem. Furthermore, we prove that global energy eigenstates are locally close to diagonal in the local energy eigenbasis, which constitutes a part of the eigenstate thermalization hypothesis that is valid regardless of the integrability of the model.

List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

Summary

We haven't generated a summary for this paper yet.

Ai Generate Text Spark Streamline Icon: https://streamlinehq.com

Paper Prompts

Sign up for free to create and run prompts on this paper using GPT-5.

Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.

Don't miss out on important new AI/ML research

See which papers are being discussed right now on X, Reddit, and more:

“Emergent Mind helps me see which AI papers have caught fire online.”

Philip

Philip

Creator, AI Explained on YouTube