arxiv: 2605.10688 · v1 · submitted 2026-05-11 · 💻 cs.LG · eess.SP

Recognition: no theorem link

DANCE: Detect and Classify Events in EEG

Jarod L\'evy , Hubert Banville , J\'er\'emy Rapin , Jean-Remi King , Thomas Moreau , St\'ephane d'Ascoli

Authors on Pith no claims yet

Pith reviewed 2026-05-12 05:27 UTC · model grok-4.3

classification 💻 cs.LG eess.SP

keywords EEGevent detectiondeep learningset predictionseizure monitoringBCIneural decoding

0 comments

The pith

One neural network jointly detects and classifies EEG events of any duration from raw unaligned recordings.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

DANCE reframes EEG event identification as a set-prediction task so that a single model can both find when events occur and label their type from continuous recordings that have no pre-marked starting points. Current methods mostly classify short windows that are already aligned to event onsets, which works in controlled studies but fails for real-world monitoring where those alignments are missing. The model was tested on ten separate datasets that include short cognitive events, longer clinical ones like seizures, and BCI signals, beating previous approaches in most cases. It reaches top performance for seizure detection and equals the results of models that do get onset information for BCI tasks. This opens a path to decoding systems that run directly on streaming brain data without extra preprocessing steps.

Core claim

DANCE frames neural decoding as a set-prediction problem to jointly detect and classify events directly from raw, unaligned signals, achieving superior performance on diverse tasks including a new state of the art in seizure monitoring while matching the accuracy of onset-informed models for BCI tasks.

What carries the argument

Set prediction, which lets the model output an unordered collection of detected events each tagged with class and timing, instead of classifying fixed aligned segments.

If this is right

Outperforms existing methods on a broad range of cognitive, clinical, and BCI tasks
Establishes a new state of the art in seizure monitoring
Matches the accuracy of onset-informed models for BCI tasks
Advances toward end-to-end asynchronous neural decoding from continuous signals

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This could allow automatic analysis of long clinical recordings without manual event marking
The set-prediction idea might transfer to event detection in other continuous time series such as ECG or audio
Future tests could check performance on multi-channel or noisy ambulatory recordings

Load-bearing premise

That treating event identification as predicting unordered sets of class-and-timing pairs suffices to handle variable event durations and lack of alignment in raw EEG data.

What would settle it

A new EEG dataset with many overlapping or rare events where the model misses detections or misclassifies more often than aligned-window baselines.

read the original abstract

Event identification in continuous neural recordings is a critical task in neuroscience. Decoding in EEG is dominated by classifying windows aligned to known event onsets. However, while available in controlled experiments, such onsets are absent in continuous real-world monitoring. Here, we introduce DANCE, a deep learning pipeline that frames neural decoding as a set-prediction problem and jointly detects and classifies events directly from raw, unaligned signals. Evaluated separately on ten datasets curated from the literature with a wide variety of event types (ranging from milliseconds to minutes in duration), our model outperforms existing methods on a broad range of cognitive, clinical and BCI tasks. This single architecture establishes a new state of the art in the competitive task of seizure monitoring and matches the accuracy of onset-informed models for BCI tasks. Overall, our method marks a step towards end-to-end asynchronous neural decoding models

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

DANCE reframes EEG event detection as set prediction to skip alignment, but the ms-to-minutes duration range looks like an unaddressed architectural risk.

read the letter

The core idea here is to treat event detection and classification in raw EEG as a set-prediction problem rather than the usual aligned-window classification. That shift lets the model work on unaligned continuous recordings, which matters for real-world monitoring where onsets are not given. The paper tests this single pipeline across ten literature-curated datasets that span cognitive tasks, clinical events, and BCI, claiming it beats prior methods on seizure detection and reaches parity with onset-informed models on BCI tasks. If the numbers hold, that breadth is useful because most current approaches stay narrow or require alignment cues. The evaluation on external datasets avoids obvious circularity and the citation pattern looks standard for the area. The soft spot is the duration range. Events from milliseconds to minutes differ by orders of magnitude, yet standard set-prediction setups with learned queries and bipartite matching run at fixed temporal granularity. Without shown multi-scale encoders, hierarchical queries, or duration-aware losses, short events risk being lost to pooling while long ones may exceed context or produce unstable matches. The abstract and available description give no equations, diagrams, or ablations that resolve this, so the single-architecture claim across all tasks rests on an assumption that still needs evidence. This work is aimed at people building asynchronous neural decoders for clinical or assistive use. It is coherent enough on its own terms to deserve a serious referee, even if the duration robustness section will need strengthening. I would send it to review rather than desk reject, but flag the scale issue for the authors to address with concrete architecture details and targeted ablations.

Referee Report

1 major / 2 minor

Summary. The paper introduces DANCE, a deep learning pipeline that frames neural decoding as a set-prediction problem to jointly detect and classify events directly from raw, unaligned EEG signals. It evaluates the approach separately on ten literature-curated datasets spanning cognitive, clinical, and BCI tasks with event durations from milliseconds to minutes, claiming outperformance over existing methods, new state-of-the-art results in seizure monitoring, and accuracy matching onset-informed models for BCI tasks.

Significance. If the results hold, the work would be significant for enabling end-to-end asynchronous decoding without reliance on event onsets, which are typically unavailable in real-world continuous monitoring. A single architecture succeeding across such diverse event types and durations could advance both clinical applications like seizure detection and practical BCI systems.

major comments (1)

[§3] The description of the set-prediction architecture (learned queries and bipartite matching) provides no explicit multi-scale encoders, hierarchical queries, duration-conditioned losses, or ablations to handle event durations spanning milliseconds to minutes. Standard fixed-granularity set prediction risks losing short events to pooling or failing on long events due to context limits; this assumption is load-bearing for the broad outperformance and SOTA claims across all tasks.

minor comments (2)

[Abstract] The abstract asserts outperformance and SOTA results but includes no quantitative metrics, baseline details, statistical tests, or validation procedures, which weakens the ability to assess the central empirical claims.
[§4] The manuscript should clarify the exact references and characteristics of the ten curated datasets to allow reproducibility and assessment of task diversity.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback. We address the single major comment point-by-point below, clarifying the architecture's handling of variable event durations while committing to improved exposition in the revised manuscript.

read point-by-point responses

Referee: [§3] The description of the set-prediction architecture (learned queries and bipartite matching) provides no explicit multi-scale encoders, hierarchical queries, duration-conditioned losses, or ablations to handle event durations spanning milliseconds to minutes. Standard fixed-granularity set prediction risks losing short events to pooling or failing on long events due to context limits; this assumption is load-bearing for the broad outperformance and SOTA claims across all tasks.

Authors: We appreciate the referee's emphasis on temporal scale robustness. Our architecture does not employ fixed-granularity pooling; the backbone is a temporal convolutional network whose feature maps preserve sufficient resolution for millisecond-scale events, while the transformer encoder with learned queries uses self- and cross-attention over the entire sequence to capture long-range context for minute-scale events. Each query independently predicts start time, duration, and class, and the bipartite matching loss (Hungarian algorithm) optimizes assignments without requiring duration-conditioned losses or hierarchical queries—the matching naturally accommodates variable durations by comparing predicted intervals to ground-truth intervals. Although dedicated ablations isolating duration handling were not included, the consistent superiority across ten datasets spanning ERP components (ms), motor imagery (hundreds of ms), and seizures (minutes) provides empirical support. We will revise §3 to explicitly describe the implicit multi-resolution properties of the attention mechanism and the interval-based prediction, thereby strengthening the exposition without altering the core claims. revision: partial

Circularity Check

0 steps flagged

No circularity; derivation is self-contained and externally evaluated

full rationale

The paper introduces DANCE as a novel pipeline framing neural decoding as a set-prediction problem for joint detection/classification from raw unaligned EEG signals. It is evaluated on ten external literature-curated datasets spanning cognitive, clinical, and BCI tasks, with claims of outperformance and SOTA on seizure monitoring. No equations, self-citations, or derivation steps are described that reduce any prediction or result to fitted inputs by construction, self-definitional loops, or load-bearing self-citations. The handling of diverse event durations is presented as an empirical capability of the architecture, tested against benchmarks rather than assumed via internal fitting or renaming. This matches the default expectation of a non-circular ML methods paper with independent validation.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard deep learning assumptions about signal informativeness and the viability of set-prediction for variable-duration events; no specific free parameters or invented entities are detailed in the abstract.

free parameters (1)

Neural network weights and hyperparameters
Fitted during training on the EEG datasets to enable the set-prediction task.

axioms (1)

domain assumption Raw EEG signals contain sufficient information to jointly detect and classify events without onset alignment
Invoked to justify the set-prediction approach over traditional aligned-window methods.

pith-pipeline@v0.9.0 · 5468 in / 1277 out tokens · 99047 ms · 2026-05-12T05:27:10.405117+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

117 extracted references · 117 canonical work pages · 2 internal anchors

[1]

Sparrkulee: A speech-evoked auditory response repository from ku leuven, containing the

Accou, Bernd and Bollens, Lies and Gillis, Marlies and Verheijen, Wendy and Van Hamme, Hugo and Francart, Tom , journal=. Sparrkulee: A speech-evoked auditory response repository from ku leuven, containing the. 2024 , publisher=

work page 2024
[2]

Sensors , volume=

A brain-controlled and user-centered intelligent wheelchair: A feasibility study , author=. Sensors , volume=. 2024 , publisher=

work page 2024
[3]

and Li, S

Wu, H. and Li, S. and Wu, D. , journal =. Motor imagery classification for asynchronous

work page
[4]

Frontiers in Human Neuroscience , volume=

Toward an adapted neurofeedback for post-stroke motor rehabilitation: state of the art and perspectives , author=. Frontiers in Human Neuroscience , volume=. 2022 , publisher=

work page 2022
[5]

REVE: A Foundation Model for

El Ouahidi, Yassine and Lys, Jonathan and Th. REVE: A Foundation Model for. Advances in Neural Information Processing Systems , year =

work page
[6]

Deep learning for electroencephalogram (

Craik, Alexander and He, Yongtian and Contreras-Vidal, Jose L , journal=. Deep learning for electroencephalogram (. 2019 , publisher=

work page 2019
[7]

Current Biology , volume=

Electrophysiological correlates of semantic dissimilarity reflect the comprehension of natural, narrative speech , author=. Current Biology , volume=. 2018 , publisher=

work page 2018
[9]

The temple university hospital

Obeid, Iyad and Picone, Joseph , journal=. The temple university hospital. 2016 , publisher=

work page 2016
[10]

Nature , volume=

High-performance brain-to-text communication via handwriting , author=. Nature , volume=. 2021 , publisher=

work page 2021
[11]

Nature , volume=

A high-performance neuroprosthesis for speech decoding and avatar control , author=. Nature , volume=. 2023 , publisher=

work page 2023
[12]

Advances in Neural Information Processing Systems (NeurIPS) , year=

emg2qwerty: A large dataset with baselines for touch typing using surface electromyography , author=. Advances in Neural Information Processing Systems (NeurIPS) , year=

work page
[13]

preprint arXiv:2506.12356 , year=

SplashNet: Split-and-Share Encoders for Accurate and Efficient Typing with Surface Electromyography , author=. preprint arXiv:2506.12356 , year=

work page arXiv
[14]

Nature Machine Intelligence , volume=

Decoding speech perception from non-invasive brain recordings , author=. Nature Machine Intelligence , volume=. 2023 , publisher=

work page 2023
[15]

Review of the

Tangermann, Michael and M. Review of the. Frontiers in neuroscience , volume=. 2012 , publisher=

work page 2012
[16]

Autocalibration and recurrent adaptation: Towards a plug and play online

Faller, Josef and Vidaurre, Carmen and Solis-Escalante, Teodoro and Neuper, Christa and Scherer, Reinhold , journal=. Autocalibration and recurrent adaptation: Towards a plug and play online. 2012 , publisher=

work page 2012
[17]

Frontiers in neuroinformatics , volume=

The temple university hospital seizure detection corpus , author=. Frontiers in neuroinformatics , volume=. 2018 , publisher=

work page 2018
[18]

Congedo, Marco and Goyat, Matthieu and Tarrin, Nicolas and Ionescu, Gelu and Varnet, L. ". International Brain-Computer Interface Conference (BCI) , year=

work page
[19]

2019 , institution=

Korczowski, Louis and Ostaschenko, Ekaterina and Andreev, Anton and Cattan, Gr. 2019 , institution=

work page 2019
[21]

The temple university artifact corpus: An annotated corpus of

Hamid, Ahmed and Gagliano, Katherine and Rahman, Safwanur and Tulin, Nikita and Tchiong, Vincent and Obeid, Iyad and Picone, Joseph , booktitle=. The temple university artifact corpus: An annotated corpus of. 2020 , organization=

work page 2020
[22]

IEEE reviews in biomedical engineering , volume=

Clinical applications of brain-computer interfaces: current state and future prospects , author=. IEEE reviews in biomedical engineering , volume=. 2009 , publisher=

work page 2009
[23]

SzCORE as a benchmark: report from the seizure detection challenge at the 2025

Dan, Jonathan and Shahbazinia, Amirhossein and Kechris, Christodoulos and Atienza, David , journal=. SzCORE as a benchmark: report from the seizure detection challenge at the 2025

work page 2025
[24]

European Conference on Computer Vision (ECCV) , year=

End-to-end object detection with transformers , author=. European Conference on Computer Vision (ECCV) , year=

work page
[25]

International Conference on Learning Representations (ICLR) , year=

Self-supervised graph neural networks for improved electroencephalographic seizure analysis , author=. International Conference on Learning Representations (ICLR) , year=

work page
[26]

Conference on health, inference, and learning , year=

Modeling multivariate biosignals with graph neural networks and structured state space models , author=. Conference on health, inference, and learning , year=

work page
[28]

Learning from heterogeneous

Saeed, Aaqib and Grangier, David and Pietquin, Olivier and Zeghidour, Neil , booktitle=. Learning from heterogeneous. 2021 , organization=

work page 2021
[30]

2019 , publisher=

Chambon, Stanislas and Thorey, Valentin and Arnal, Pierrick J and Mignot, Emmanuel and Gramfort, Alexandre , journal=. 2019 , publisher=

work page 2019
[31]

IEEE Transactions on Biomedical Engineering , volume=

Avoiding post-processing with event-based detection in biomedical signals , author=. IEEE Transactions on Biomedical Engineering , volume=

work page
[33]

Kuhn, H. W. , year = 1955, journal =. The

work page 1955
[34]

Nature methods , volume=

Metrics reloaded: recommendations for image analysis validation , author=. Nature methods , volume=. 2024 , publisher=

work page 2024
[35]

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=

Generalized Intersection over Union: A Metric and a Loss for Bounding Box Regression , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=

work page
[36]

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=

You only look once: Unified, real-time object detection , author=. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=

work page
[37]

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=

Rich feature hierarchies for accurate object detection and semantic segmentation , author=. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=

work page
[38]

Advances in Neural Information Processing Systems (NeurIPS) , year=

Faster r-cnn: Towards real-time object detection with region proposal networks , author=. Advances in Neural Information Processing Systems (NeurIPS) , year=

work page
[39]

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=

Dn-detr: Accelerate detr training by introducing query denoising , author=. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year=

work page
[40]

International Conference on Machine Learning (ICML) , year=

A deep learning approach for the segmentation of electroencephalography data in eye tracking applications , author=. International Conference on Machine Learning (ICML) , year=

work page
[42]

Zhao, Yian and Lv, Wenyu and Xu, Shangliang and Wei, Jinman and Wang, Guanzhong and Dang, Qingqing and Liu, Yi and Chen, Jie , booktitle=

work page
[43]

International Conference on Machine Learning (ICML) , year=

Perceiver: General perception with iterative attention , author=. International Conference on Machine Learning (ICML) , year=

work page
[44]

SciPy , volume=

Statsmodels: econometric and statistical modeling with python , author=. SciPy , volume=

work page
[45]

Deep Learning sur les signaux

L. Deep Learning sur les signaux. XXIX. 2025 , address =

work page 2025
[46]

Journal of Machine Learning Research (JMLR) , volume=

Scikit-learn: Machine learning in Python , author=. Journal of Machine Learning Research (JMLR) , volume=. 2011 , publisher=

work page 2011
[47]

International Conference on Learning Representations (ICLR) , year=

Decoupled Weight Decay Regularization , author=. International Conference on Learning Representations (ICLR) , year=

work page
[48]

Artificial intelligence and machine learning for multi-domain operations applications , year=

Super-convergence: Very fast training of neural networks using large learning rates , author=. Artificial intelligence and machine learning for multi-domain operations applications , year=

work page
[49]

2018 , publisher=

Lawhern, Vernon J and Solon, Amelia J and Waytowich, Nicholas R and Gordon, Stephen M and Hung, Chou P and Lance, Brent J , journal=. 2018 , publisher=

work page 2018
[50]

Physics-informed attention temporal convolutional network for

Altaheri, Hamdi and Muhammad, Ghulam and Alsulaiman, Mansour , journal=. Physics-informed attention temporal convolutional network for. 2023 , publisher=

work page 2023
[52]

NPJ digital medicine , volume=

U-Sleep: resilient high-frequency sleep staging , author=. NPJ digital medicine , volume=. 2021 , publisher=

work page 2021
[53]

Jiquan Wang and Sha Zhao and Zhiling Luo and Yangxuan Zhou and Haiteng Jiang and Shijian Li and Tao Li and Gang Pan , booktitle=

work page
[54]

Deep learning with convolutional neural networks for

Schirrmeister, Robin Tibor and Springenberg, Jost Tobias and Fiederer, Lukas Dominique Josef and Glasstetter, Martin and Eggensperger, Katharina and Tangermann, Michael and Hutter, Frank and Burgard, Wolfram and Ball, Tonio , journal=. Deep learning with convolutional neural networks for. 2017 , publisher=

work page 2017
[55]

Large Brain Model for Learning Generic Representations with Tremendous

Weibang Jiang and Liming Zhao and Bao-liang Lu , booktitle=. Large Brain Model for Learning Generic Representations with Tremendous

work page
[56]

Weibang Jiang and Yansen Wang and Bao-liang Lu and Dongsheng Li , booktitle=. Neuro

work page
[57]

Wang, Shuo and Xia, Chunlong and Lv, Feng and others , booktitle=

work page
[58]

European Conference on Computer Vision (ECCV) , year=

Actionformer: Localizing moments of actions with transformers , author=. European Conference on Computer Vision (ECCV) , year=

work page
[59]

Zong, Zhuofan and Song, Guanglu and Liu, Yu , booktitle=

work page
[60]

Automated multi-class seizure-type classification system using

Abirami, S and Kathiravan, M and Yuvaraj, Rajamanickam and Menon, Ramshekhar N and Thomas, John and Karthick, PA and Prince, A Amalin and Ronickom, Jac Fredo Agastinose and others , journal=. Automated multi-class seizure-type classification system using. 2024 , publisher=

work page 2024
[61]

Automated seizure detection using transformer models on multi-channel

Zhu, Yuanda and Wang, May D , booktitle=. Automated seizure detection using transformer models on multi-channel. 2023 , organization=

work page 2023
[62]

2022 , publisher=

Thuwajit, Punnawish and Rangpong, Phurin and Sawangjai, Phattarapong and Autthasan, Phairot and Chaisaen, Rattanaphon and Banluesombatkul, Nannapas and Boonchit, Puttaranun and Tatsaringkansakul, Nattasate and Sudhawiyangkul, Thapanun and Wilaiprasitporn, Theerawit , journal=. 2022 , publisher=

work page 2022
[65]

Microsoft

Lin, Tsung-Yi and Maire, Michael and Belongie, Serge and Hays, James and Perona, Pietro and Ramanan, Deva and Doll. Microsoft. European Conference on Computer Vision (ECCV) , year=

work page
[66]

International Conference on Machine learning (ICML) , year=

Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks , author=. International Conference on Machine learning (ICML) , year=

work page
[67]

Automated multi-class seizure-type classification system using EEG signals and machine learning algorithms

S Abirami, M Kathiravan, Rajamanickam Yuvaraj, Ramshekhar N Menon, John Thomas, PA Karthick, A Amalin Prince, Jac Fredo Agastinose Ronickom, et al. Automated multi-class seizure-type classification system using EEG signals and machine learning algorithms. IEEE Access, 12: 0 136524--136541, 2024

work page 2024
[68]

Sparrkulee: A speech-evoked auditory response repository from ku leuven, containing the EEG of 85 participants

Bernd Accou, Lies Bollens, Marlies Gillis, Wendy Verheijen, Hugo Van Hamme, and Tom Francart. Sparrkulee: A speech-evoked auditory response repository from ku leuven, containing the EEG of 85 participants. Data, 9 0 (8): 0 94, 2024

work page 2024
[69]

Physics-informed attention temporal convolutional network for EEG -based motor imagery classification

Hamdi Altaheri, Ghulam Muhammad, and Mansour Alsulaiman. Physics-informed attention temporal convolutional network for EEG -based motor imagery classification. IEEE Transactions on Industrial Informatics, 19 0 (2): 0 2249--2258, 2023

work page 2023
[70]

Scaling laws for decoding images from brain activity.arXiv preprint arXiv:2501.15322, 2025

Hubert Banville, Yohann Benchetrit, St \'e phane d'Ascoli, J \'e r \'e my Rapin, and Jean-R \'e mi King. Scaling laws for decoding images from brain activity. preprint arXiv:2501.15322, 2025

work page arXiv 2025
[71]

Electrophysiological correlates of semantic dissimilarity reflect the comprehension of natural, narrative speech

Michael P Broderick, Andrew J Anderson, Giovanni M Di Liberto, Michael J Crosse, and Edmund C Lalor. Electrophysiological correlates of semantic dissimilarity reflect the comprehension of natural, narrative speech. Current Biology, 28 0 (5): 0 803--809, 2018

work page 2018
[72]

End-to-end object detection with transformers

Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko. End-to-end object detection with transformers. In European Conference on Computer Vision (ECCV). Springer, 2020

work page 2020
[73]

DOSED : A deep learning approach to detect multiple sleep micro-events in EEG signal

Stanislas Chambon, Valentin Thorey, Pierrick J Arnal, Emmanuel Mignot, and Alexandre Gramfort. DOSED : A deep learning approach to detect multiple sleep micro-events in EEG signal. Journal of neuroscience methods, 321: 0 64--78, 2019

work page 2019
[74]

Toward interpretable evaluation measures for time series segmentation

F \'e lix Chavelli, Paul Boniol, and Micha \"e l Thomazo. Toward interpretable evaluation measures for time series segmentation. preprint arXiv:2510.23261, 2025

work page arXiv 2025
[75]

Subject-aware contrastive learning for biosignals

Joseph Y Cheng, Hanlin Goh, Kaan Dogrusoz, Oncel Tuzel, and Erdrin Azemi. Subject-aware contrastive learning for biosignals. preprint arXiv:2007.04871, 2020

work page arXiv 2007
[76]

The largest EEG -based BCI reproducibility study for open science: the moabb benchmark

Sylvain Chevallier, Igor Carrara, Bruno Aristimunha, Pierre Guetschel, Sara Sedlar, Bruna Lopes, S \'e bastien Velut, Salim Khazem, and Thomas Moreau. The largest EEG -based BCI reproducibility study for open science: the moabb benchmark. preprint arXiv:2404.15319, 2024

work page arXiv 2024
[77]

e l Acquadro, and Christian Jutten

Marco Congedo, Matthieu Goyat, Nicolas Tarrin, Gelu Ionescu, L \'e o Varnet, Bertrand Rivet, Ronald Phlypo, Nisrine Jrad, Micha \"e l Acquadro, and Christian Jutten. " Brain Invaders ": a prototype of an open-source P300 -based video game working with the openvibe platform. In International Brain-Computer Interface Conference (BCI), 2011

work page 2011
[78]

Deep learning for electroencephalogram ( EEG ) classification tasks: a review

Alexander Craik, Yongtian He, and Jose L Contreras-Vidal. Deep learning for electroencephalogram ( EEG ) classification tasks: a review. Journal of neural engineering, 16 0 (3): 0 031001, 2019

work page 2019
[79]

Szcore as a benchmark: report from the seizure detection challenge at the 2025 AI in epilepsy and neurological disorders conference

Jonathan Dan, Amirhossein Shahbazinia, Christodoulos Kechris, and David Atienza. Szcore as a benchmark: report from the seizure detection challenge at the 2025 AI in epilepsy and neurological disorders conference. preprint arXiv:2505.18191, 2025

work page arXiv 2025
[80]

Decoding speech perception from non-invasive brain recordings

Alexandre D \'e fossez, Charlotte Caucheteux, J \'e r \'e my Rapin, Ori Kabeli, and Jean-R \'e mi King. Decoding speech perception from non-invasive brain recordings. Nature Machine Intelligence, 5 0 (10): 0 1097--1107, 2023

work page 2023
[81]

Reve: A foundation model for EEG -- adapting to any setup with large-scale pretraining on 25,000 subjects

Yassine El Ouahidi, Jonathan Lys, Philipp Th \"o lke, Nicolas Farrugia, Bastien Pasdeloup, Vincent Gripon, Karim Jerbi, and Giulia Lioi. Reve: A foundation model for EEG -- adapting to any setup with large-scale pretraining on 25,000 subjects. In Advances in Neural Information Processing Systems, 2025

work page 2025
[82]

Autocalibration and recurrent adaptation: Towards a plug and play online ERD - BCI

Josef Faller, Carmen Vidaurre, Teodoro Solis-Escalante, Christa Neuper, and Reinhold Scherer. Autocalibration and recurrent adaptation: Towards a plug and play online ERD - BCI . IEEE Transactions on Neural Systems and Rehabilitation Engineering, 20 0 (3): 0 313--319, 2012

work page 2012
[83]

Rich feature hierarchies for accurate object detection and semantic segmentation

Ross Girshick, Jeff Donahue, Trevor Darrell, and Jitendra Malik. Rich feature hierarchies for accurate object detection and semantic segmentation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2014

work page 2014
[84]

The temple university artifact corpus: An annotated corpus of EEG artifacts

Ahmed Hamid, Katherine Gagliano, Safwanur Rahman, Nikita Tulin, Vincent Tchiong, Iyad Obeid, and Joseph Picone. The temple university artifact corpus: An annotated corpus of EEG artifacts. In Signal Processing in Medicine and Biology Symposium (SPMB). IEEE, 2020

work page 2020
[85]

Distilling the Knowledge in a Neural Network

Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. Distilling the knowledge in a neural network. preprint arXiv:1503.02531, 2015

work page internal anchor Pith review Pith/arXiv arXiv 2015
[86]

Perceiver: General perception with iterative attention

Andrew Jaegle, Felix Gimeno, Andy Brock, Oriol Vinyals, Andrew Zisserman, and Joao Carreira. Perceiver: General perception with iterative attention. In International Conference on Machine Learning (ICML). PMLR, 2021

work page 2021
[87]

Large brain model for learning generic representations with tremendous EEG data in BCI

Weibang Jiang, Liming Zhao, and Bao liang Lu. Large brain model for learning generic representations with tremendous EEG data in BCI . In International Conference on Learning Representations (ICLR), 2024

work page 2024
[88]

Brain Invaders calibration-less P300 -based BCI using dry EEG electrodes dataset (bi2014a)

Louis Korczowski, Ekaterina Ostaschenko, Anton Andreev, Gr \'e goire Cattan, Pedro Luiz Coelho Rodrigues, Violette Gautheret, and Marco Congedo. Brain Invaders calibration-less P300 -based BCI using dry EEG electrodes dataset (bi2014a). Technical report, GIPSA-lab, 2019 a

work page 2019
[89]

Brain Invaders solo versus collaboration: Multi-user P300 -based brain-computer interface dataset (bi2014b)

Louis Korczowski, Ekaterina Ostaschenko, Anton Andreev, Gr \'e goire Cattan, Pedro Luiz Coelho Rodrigues, Violette Gautheret, and Marco Congedo. Brain Invaders solo versus collaboration: Multi-user P300 -based brain-computer interface dataset (bi2014b). Technical report, GIPSA-lab, 2019 b

work page 2019

Showing first 80 references.