QiT: Quantum-Inspired Transformer for Visual Recognition Task

Abstract

Quantum machine learning offers a compelling representational perspective: angle-encoded states inhabit Hilbert spaces in which periodic similarities and interactions can be expressed naturally. Realizing this perspective for visual recognition remains difficult, however, because present quantum neural networks are constrained by limited qubit counts, costly circuit simulation and measurement, noise, and unstable optimization on noisy intermediate-scale quantum devices. We investigate whether useful structural ideas from quantum models can instead be realized as scalable classical Transformer operations. We introduce QiT, a Quantum-inspired Transformer for vision tasks with three components: (i) angle-inspired encoding that maps image tokens to learned trigonometric Hilbert-space features analogous to quantum rotation-based state encoding; (ii) self-attention over these periodic features, inducing a classical cosine kernel approximated to quantum fidelity kernels; and (iii) gated multiplicative emulation, a trainable classical surrogate for interaction terms found in variational circuits. All components are differentiable tensor operations, so QiT claims neither quantum computation nor quantum speedup and retains the 𝒪(N²D) attention complexity of a standard Vision Transformer. Across image-classification benchmarks, QiT is competitive with a matched classical Transformer while avoiding the severe runtime cost observed for a small simulated quantum Transformer. QiT-B reaches 78.3\% ImageNet-1K top-1 accuracy with 45.7M parameters and 11.5 GFLOPs. These results position QiT as a scalable baseline for isolating and evaluating quantum-motivated inductive biases in visual recognition.

Transformer-inspired architectures make machine-learned electron densities accurate and transferable enough to bypass self-consistency
Jeheon Woo
arXiv29 Oct 2026PreprintarXiv licenceFull textPhysical chemistry Computational physicsarXiv
Few-shot recognition of lane-changing intention in connected vehicle environment: A meta-transformer approach
Xiaoqian Liu, Wenxin Jiang, Yining Ren
PLOS One29 Oct 2026Peer-reviewedCC BYFull textNeuroscience General mathematicsPLOSEurope PMCPMC
ConCA: Concentration-Aware Channel Attention for Fine-Grained Visual Recognition
Yu-Sheng Liu, Yu-Chen Tung
arXiv31 Aug 2026PreprintarXiv licenceFull textHigh-energy & particle physicsarXiv
Optical measurement is almost quantum: Quantum inspired universal optical bounds
Daniel Ish, Chase T. Ellis
arXiv30 Sep 2026PreprintarXiv licenceFull textOptics & photonics Quantum physics & quantum informationarXiv
Variational study of the magnetization plateaus in the spin-1/2 kagome Heisenberg antiferromagnet: An approach from vision transformer neural quantum states
Andreas Raikos, Sylvain Capponi, Fabien Alet
Physical Review B29 Sep 2026Peer-reviewedLicence not statedFull textCondensed matterHAL

Cite

BibTeX RIS

BibTeX

@misc{patro2026quantuminspired,
  title = {{QiT: Quantum-Inspired Transformer for Visual Recognition Task}},
  author = {Badri N. Patro and Vijay Agneeswaran},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.17789},
  eprint = {2609.17789},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.17789},
  note = {CC BY-NC-SA 4.0},
}

RIS

TY  - UNPB
TI  - QiT: Quantum-Inspired Transformer for Visual Recognition Task
AU  - Badri N. Patro
AU  - Vijay Agneeswaran
PY  - 2026
DA  - 2026/09/15
PB  - arXiv
DO  - 10.48550/arxiv.2609.17789
UR  - https://arxiv.org/abs/2609.17789
AB  - Quantum machine learning offers a compelling representational perspective: angle-encoded states inhabit Hilbert spaces in which periodic similarities and interactions can be expressed naturally. Realizing this perspective for visual recognition remains difficult, however, because present quantum neural networks are constrained by limited qubit counts, costly circuit simulation and measurement, noise, and unstable optimization on noisy intermediate-scale quantum devices. We investigate whether useful structural ideas from quantum models can instead be realized as scalable classical Transformer operations. We introduce QiT, a Quantum-inspired Transformer for vision tasks with three components: (i) angle-inspired encoding that maps image tokens to learned trigonometric Hilbert-space features analogous to quantum rotation-based state encoding; (ii) self-attention over these periodic features, inducing a classical cosine kernel approximated to quantum fidelity kernels; and (iii) gated multiplicative emulation, a trainable classical surrogate for interaction terms found in variational circuits. All components are differentiable tensor operations, so QiT claims neither quantum computation nor quantum speedup and retains the 𝒪(N^2D) attention complexity of a standard Vision Transformer. Across image-classification benchmarks, QiT is competitive with a matched classical Transformer while avoiding the severe runtime cost observed for a small simulated quantum Transformer. QiT-B reaches 78.3\% ImageNet-1K top-1 accuracy with 45.7M parameters and 11.5 GFLOPs. These results position QiT as a scalable baseline for isolating and evaluating quantum-motivated inductive biases in visual recognition.
ER  -

Abstract

Related

Cite