Saved in:
Bibliographic Details
Main Authors: Gkikas, Stefanos, Cruz, Christian Arzate, Fang, Yu, Cao, Lu, Khan, Muhammad Umar, Kassiotis, Thomas, Giannakakis, Giorgos, Rojas, Raul Fernandez, Gomez, Randy
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.16491
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913143145365504
author Gkikas, Stefanos
Cruz, Christian Arzate
Fang, Yu
Cao, Lu
Khan, Muhammad Umar
Kassiotis, Thomas
Giannakakis, Giorgos
Rojas, Raul Fernandez
Gomez, Randy
author_facet Gkikas, Stefanos
Cruz, Christian Arzate
Fang, Yu
Cao, Lu
Khan, Muhammad Umar
Kassiotis, Thomas
Giannakakis, Giorgos
Rojas, Raul Fernandez
Gomez, Randy
contents Pain is a multifaceted and widespread phenomenon with substantial clinical and societal burden, making reliable automated assessment a critical objective. This paper presents a lightweight transformer architecture that fuses multiple fNIRS representations through a unified tokenization mechanism, enabling joint modeling of complementary signal views without requiring modality-specific adaptations or increasing architectural complexity. The proposed token-mixing strategy preserves spatial, temporal, and time-frequency characteristics by projecting heterogeneous inputs onto a shared latent representation, using a structured segmentation scheme to control the granularity of local aggregation and global interaction. The model is evaluated on the AI4Pain dataset using stacked raw waveform and power spectral density representations of fNIRS inputs. Experimental results demonstrate competitive pain recognition performance while remaining computationally compact, making the approach suitable for real-time inference on both GPU and CPU hardware.
format Preprint
id arxiv_https___arxiv_org_abs_2604_16491
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Lightweight Transformer for Pain Recognition from Brain Activity
Gkikas, Stefanos
Cruz, Christian Arzate
Fang, Yu
Cao, Lu
Khan, Muhammad Umar
Kassiotis, Thomas
Giannakakis, Giorgos
Rojas, Raul Fernandez
Gomez, Randy
Computer Vision and Pattern Recognition
Artificial Intelligence
Pain is a multifaceted and widespread phenomenon with substantial clinical and societal burden, making reliable automated assessment a critical objective. This paper presents a lightweight transformer architecture that fuses multiple fNIRS representations through a unified tokenization mechanism, enabling joint modeling of complementary signal views without requiring modality-specific adaptations or increasing architectural complexity. The proposed token-mixing strategy preserves spatial, temporal, and time-frequency characteristics by projecting heterogeneous inputs onto a shared latent representation, using a structured segmentation scheme to control the granularity of local aggregation and global interaction. The model is evaluated on the AI4Pain dataset using stacked raw waveform and power spectral density representations of fNIRS inputs. Experimental results demonstrate competitive pain recognition performance while remaining computationally compact, making the approach suitable for real-time inference on both GPU and CPU hardware.
title A Lightweight Transformer for Pain Recognition from Brain Activity
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2604.16491