YUCHA: Yet Underrated Conservative Hamiltonian Attention

This paper proposes YUCHA (Yet Underrated Conservative Hamiltonian Attention), a Transformer architecture that applies Hamiltonian dynamics and a 2nd-order Leapfrog integrator to the attention mechanism, enforcing physical energy conservation across layers to improve long-term optimization stability. Source code included. Architecture design assisted by Google Antigravity (AI).

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-07-28
DOI
https://doi.org/10.5281/zenodo.21648937
Primary Topic
Parallel Computing and Optimization Techniques
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

YUCHA: Yet Underrated Conservative Hamiltonian Attention

hyukkyu song
Zenodo (CERN European Organization for Nuclear Research)
Parallel Computing and Optimization Techniques
preprint

YUCHA: Yet Underrated Conservative Hamiltonian Attention

hyukkyu song
preprint en

Abstract

This paper proposes YUCHA (Yet Underrated Conservative Hamiltonian Attention), a Transformer architecture that applies Hamiltonian dynamics and a 2nd-order Leapfrog integrator to the attention mechanism, enforcing physical energy conservation across layers to improve long-term optimization stability. Source code included. Architecture design assisted by Google Antigravity (AI).

Zenodo (CERN European Organization for Nuclear Research)
Affordable and clean energy
Parallel Computing and Optimization Techniques
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.