Modelpublic

Premchan369/Q-TensorFormer

sourceHugging Faceapache-2.0updated 9d agoView on Hugging Face

2likes185downloads

CITATION.cff24 linesDownload Raw Back to root

1cff-version: 1.2.02title: "Q-TensorFormer: Quantum-Enhanced Tensor Network LLM Compression Engine"3message: "If you use this software in your research, please cite it using the metadata below."4authors:5  - given-names: Premchan3696    affiliation: ""7repository-code: "https://huggingface.co/Premchan369/q-tensorformer"8url: "https://huggingface.co/Premchan369/q-tensorformer"9abstract: >-10  Q-TensorFormer is a hybrid transformer architecture that compresses feed-forward11  layers using Tensor-Train (TT) decomposition and enhances token representations12  via real quantum circuits, with adaptive TT-rank scheduling guided by attention13  entropy. It achieves 50-70% parameter reduction at equivalent perplexity.14keywords:15  - tensor networks16  - quantum machine learning17  - model compression18  - transformer19  - language modeling20  - efficient deep learning21license: Apache-2.022version: 3.0.023date-released: 2026-05-0624