Pith. sign in

REVIEW 3 major objections 5 minor 47 references

EEG-PRIME: Prototype-Aligned Representation Learning with Multi-Level Conditioning for EEG Decoding

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that a single frozen EEG model, conditioned on task language and dataset identity, can decode across BCI paradigms and match calibration-based decoders on unseen datasets.

desk verdict Solid broad EEG foundation model with a serious evaluation, but the headline zero-shot result isn't interpretable until they specify what happens to the dataset embedding for unseen datasets. read the letter →

arxiv 2608.13072 v1 pith:DXWGJMR2 submitted 2026-08-13 cs.AI

Shuailei Zhang , Muyun Jiang , Wei Zhang , Jinbo Chen , Zhiwei Guo , Yong Li , Yi Ding , Cuntai Guan This is my paper · ORCID
classification cs.AI
keywords EEGfoundationmodelinstructiontuningmulti-levelconditioningprototypeclassificationzero-shottransfermotorimageryemotionrecognitionbrain-computerinterface
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper proposes EEG-PRIME, a two-stage EEG foundation model that tries to make one frozen network decode many brain-computer interface tasks at once. Its central idea is that language should steer how the model reads brain signals, not rewrite the signals themselves: a task description and a learned dataset embedding modulate a Q-Former's queries at every layer, and predictions are made by matching the resulting EEG embedding to frozen text embeddings of class labels. If true, a practical decoder could be deployed on a new person or lab without collecting calibration trials or retraining task-specific heads. On sixteen datasets the model reports the best balanced accuracy on thirteen and top-three on all, and on two held-out motor-imagery datasets the frozen model reaches balanced accuracy comparable to per-subject calibrated decoders.

What carries the argument

The load-bearing mechanism is a modified Q-Former, a compact transformer whose learnable query vectors read out EEG token sequences by cross-attention. The conditioning vector $e_{\mathrm{cond}} = e_{\mathrm{ins}} + \alpha e_d$, where $e_{\mathrm{ins}}$ is a frozen sentence embedding of the task instruction and $e_d$ is a learned dataset embedding, enters every Q-Former sublayer through Layer-wise Query Modulation (LQM): each sublayer gets per-dimension scale $\gamma(e_{\mathrm{cond}})$ and shift $\beta(e_{\mathrm{cond}})$ after layer normalization. A gradient-reversal subject classifier pushes the summary embedding $z$ to be subject-invariant, a query-diversity penalty stops query collapse, and the final prediction is $\arg\max_y \cos(z, p_y)$ against frozen text prototypes $p_y$ of class names. This replaces all task-specific heads with one semantic similarity comparison, supported by a masked-autoencoding pretraining stage with frequency-cutoff spectral augmentation.

What would settle it

Re-run the zero-shot evaluation on the two held-out motor-imagery datasets with the dataset embedding $e_d$ explicitly fixed to the zero vector, and compare to the reported 63.0% and 64.2% balanced accuracy; a large drop would show that the zero-shot results depend on a dataset signal not described in the protocol.

Watch

Extended reading notes

Core claim

EEG-PRIME's central claim is that cross-dataset, cross-subject EEG decoding can be unified without dataset-specific classifier heads: a single prototype-based head over text class embeddings, steered by task-instruction and dataset conditioning, generalizes across motor imagery, emotion recognition, ADHD detection, covert speech, and mental workload. Concretely, the paper reports winning or top-three balanced accuracy on all sixteen fine-tuning benchmarks and a statistically significant advantage over every baseline, with large effect sizes. On the two held-out datasets, the frozen model obtains 63.0% mean balanced accuracy across 60 subjects versus 62.8% for within-session CSP+LDA, and 64.2% versus 66.2–68.1% for supervised leave-one-subject-out baselines. The paper reads these numbers as evidence that zero-shot EEG decoding is feasible for paradigms with clear neural correlates, while noting in its limitations that zero-shot transfer is not yet uniform across all paradigms.

Load-bearing premise

The load-bearing premise is that a meaningful dataset-conditioning vector can be supplied for a dataset never seen during training, but the paper does not specify what $e_d$ is set to at zero-shot test time, and the ablation shows that dropping it substantially hurts frozen inference.

Editorial extensions

If this is right

  • If the result holds, new BCI datasets can be added without designing a new classifier head: only the class-name text prototypes and a task instruction are needed.
  • Zero-shot deployment becomes practical for motor imagery: a frozen model can be run on a new user's session with no calibration, at accuracy comparable to per-subject calibrated CSP+LDA.
  • Instruction conditioning in the Q-Former gives a principled place for language in EEG decoding: semantics modulate query geometry rather than being concatenated with EEG tokens, so attention is steered without distorting the neural signal.
  • The finding that scale modulation $\gamma$ drives most of the gain suggests LQM's effectiveness comes from multiplicative reshaping of normalized query states, and that shift-only variants are insufficient.
  • The Q-Former's cross-attention aligns with event-related desynchronization (correlation $r=0.83$), indicating the learned reader focuses on the same temporal windows a neurophysiologist would, supporting interpretability.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A reader extending the method would need to pin down the zero-shot dataset-embedding protocol; the paper's ablation makes this non-optional, since dropping $e_d$ cuts frozen in-domain inference from 0.4848 to 0.4242 balanced accuracy.
  • The dominance of scale over shift in LQM may transfer to other instruction-conditioned transformer decoders beyond EEG; a cheap test is ablating $\gamma$ and $\beta$ in a language-conditioned audio or image decoder.
  • Because class prototypes are frozen text embeddings, the label vocabulary's semantics should matter: using synonymous or paraphrased class names should shift zero-shot accuracy, and this could be used as a deliberate probe of the EEG–language alignment.
  • The paper's own limitation note implies that zero-shot transfer will likely need per-paradigm calibration for emotion and other tasks with weaker shared neurophysiology, so a practical rollout would pair this model with lightweight domain adaptation for those paradigms.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes EEG-PRIME, a two-stage EEG foundation model for cross-dataset multi-task decoding. In the first stage, an EEG encoder is pretrained on roughly 1,153 hours of unlabeled EEG from nine datasets using masked reconstruction with frequency-cutoff spectral augmentation. In the second stage, the frozen encoder is paired with a Q-Former whose queries are modulated by a conditioning vector that combines a frozen SBERT task-instruction embedding and a learned per-dataset embedding; a gradient-reversal adversarial loss encourages subject invariance, and a query-diversity loss prevents query collapse. Classification is performed by cosine similarity between the Q-Former output and frozen text prototypes of class labels, removing the need for dataset-specific classifier heads. The authors evaluate on sixteen datasets for task-specific fine-tuning and two held-out motor-imagery datasets for zero-shot inference, reporting gains over EEGNet, TSception, ST-Transformer, Conformer, BIOT, EEGPT, LaBraM, CBraMod, and MIRepNet, with three-seed runs, Wilcoxon tests, Cohen's d, and ablations of the text encoder, instruction, and dataset embedding.

Significance. If the results hold, EEG-PRIME contributes a genuinely useful design in the prototype-aligned unified head and the layer-wise query modulation mechanism, and the breadth of the evaluation (18 datasets, five paradigms, three seeds, statistical testing, and an honest limitation section) is a clear strength. The release of code and pretrained models is also commendable. However, the central zero-shot claim is currently not fully interpretable because the paper does not specify how the learned dataset embedding is supplied for the two held-out datasets, and the fine-tuning comparison is potentially confounded by the unstated use of downstream training splits during instruction tuning. The absence of NeuroLM, the closest language-aligned EEG foundation model, further weakens the comparative claim. These issues are substantial but appear fixable within the scope of a major revision.

major comments (3)
  1. [Section III-C1, Eq. (6); Section III-D, Eq. (15); Table IV] The zero-shot protocol does not specify what value the learned dataset embedding e_d takes for the held-out datasets Dreyer2023A and Weibo2014. Since e_cond = e_ins + alpha * e_d in Eq. (6), and Table IV shows that removing e_d degrades in-domain direct inference from 0.4848 to 0.4242 balanced accuracy, the zero-shot results in Section V.B cannot be interpreted without knowing the test-time e_d. If e_d is set to zero for unseen datasets, the model is evaluated in a configuration that Table IV identifies as substantially degraded; if a non-zero value is used (e.g., the mean of trained dataset embeddings), that choice constitutes a dataset-level prior and must be disclosed and ablated. Please state the exact procedure used at zero-shot inference and, if a non-zero default is used, ablate that choice.
  2. [Section IV.A and Section III-D] The manuscript does not state which datasets are used in the instruction-tuning stage, but the sentence in Section IV.A that the two held-out datasets are 'not seen during instruction tuning or adaptation' strongly implies that the sixteen downstream datasets are used for instruction tuning. If instruction tuning is performed on the training splits of those same sixteen datasets, then the task-specific fine-tuning results in Table II compare EEG-PRIME, which has already received a supervised multi-dataset instruction-tuning pass on the target training data, against baselines that are fine-tuned directly from their pretrained checkpoints. This would confound the fine-tuning comparison. Please clarify the instruction-tuning data, and if the downstream training splits are used, add an equivalent multi-task instruction-tuning stage for all baselines or restrict instruction tuning to the pretraining corpus and re-evaluate.
  3. [Section IV.B, Section II.B] The baseline list omits NeuroLM (reference [5]), which is the closest prior work to the proposed method: it is a language-aligned EEG foundation model that performs instruction tuning and would be the most direct comparison for the paper's central claims. Without NeuroLM, the statement that EEG-PRIME achieves 'consistent improvements over ... prior EEG foundation models' is not fully supported. Please add NeuroLM to the fine-tuning and zero-shot comparisons where feasible, or provide a concrete justification for its exclusion (for example, unavailability of a pretrained checkpoint), and update Tables II and III accordingly.
minor comments (5)
  1. [Section IV.A, Section V.B] The Dreyer2023A dataset is cited as reference [29] in the dataset list but as reference [41] in the zero-shot section; please cite the correct dataset release consistently.
  2. [Table IV] The rows of Table IV are ambiguous because the checkmark placement does not clearly indicate which variant corresponds to the 'default instruction' condition and which corresponds to the 'no dataset embedding' condition; please label the rows explicitly.
  3. [Eq. (5)] The pretraining loss ell_mask(Encoder_theta, X_tilde, m) is not formally defined; please specify the exact mean squared error computation over masked token positions.
  4. [Abstract and Table I] The abstract and Section V.A describe the evaluation as being under 'cross-subject settings,' but Table I uses cross-trial splits for SEED, SEED-IV, SEED-V, and SEED-VII; please qualify the claim to 'cross-subject and cross-trial settings' where appropriate.
  5. [Eq. (1) and Section III-C2] The problem formulation writes the model as f_theta(X; d, s), but the subject identity s is never provided as an input at inference; subject conditioning is implemented only through the gradient-reversal adversarial loss in Section III-C2. Please align the notation with the actual mechanism.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the core derivation (masked pretraining, frozen text prototypes, held-out zero-shot evaluation) is self-contained and not forced by fitted parameters or self-citations.

full rationale

The paper's central claims are not circular. The two zero-shot datasets (Dreyer2023A, Weibo2014) are explicitly excluded from all training and fine-tuning stages, so their reported balanced accuracies are genuine hold-out evaluations rather than fitted inputs relabeled as predictions. The prototype classifier uses frozen Sentence-BERT text embeddings of class labels as anchors, and Eq. (15) computes cosine similarity against those external text prototypes, so the label space is not derived from or fitted to the EEG data. The masked-reconstruction pretraining objective in Eq. (5) is self-supervised and does not presuppose any downstream label. The only ambiguity is the test-time value of the per-dataset embedding e_d for unseen datasets, and Table IV shows that this embedding materially affects in-domain direct inference; however, the paper's silence on this protocol detail is a reproducibility gap, not a circular reduction, because no fitted parameter is being presented as a prediction and the reported zero-shot numbers do not reduce to a fitted quantity by construction. Self-citations such as [45]-[47] are used only as supporting references for standard neurophysiological patterns and baseline context, not as load-bearing justifications for the paper's main results. Overall, the derivation chain is independent of its own outputs and no specific equation reduces to its inputs.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central claim relies on several domain assumptions: text class names share a transferable semantic structure with EEG discriminative patterns; the 65-channel 10-10 montage interpolation preserves task-relevant information; and gradient reversal removes subject identity without removing task information. The learned dataset embeddings and scale alpha are fitted on the instruction-tuning data, and several hyperparameters (omega_1, omega_2, lambda, mask ratio) are chosen by hand.

free parameters (6)
  • dataset embedding e_d = per-dataset learned vector (dimension D = 768 assumed from SBERT)
    Learnable per-dataset embedding that captures acquisition-specific distribution; ablation shows it is critical for direct inference.
  • scale alpha = learned scalar, initialized to 1
    Fuses dataset embedding with task instruction; learned jointly during instruction tuning.
  • diversity loss weight omega_1 = 1e-3
    Chosen by hand; balances query diversity regularization term.
  • subject loss weight omega_2 = 0.5
    Chosen by hand; ramped in over 8 epochs with GRL.
  • GRL scale lambda = 0.03
    Chosen by hand; controls gradient reversal strength.
  • mask ratio r = 0.5
    Pretraining token masking ratio.
assumptions (4)
  • domain assumption Text class labels, when embedded by a frozen sentence encoder, form a semantic space whose geometry is predictive of EEG discriminative structure across datasets
    This is the core premise of prototype-aligned classification (Section III-C3); if false, zero-shot and multi-task classification lose their anchor.
  • domain assumption Interpolating arbitrary EEG montages onto a fixed 65-channel 10-10 layout preserves the information needed for decoding
    Used in EEG unification (Section III-B-a) and preprocessing (Section IV-C); lossy interpolation may discard task-relevant spatial detail.
  • domain assumption Gradient reversal removes subject identity from the representation without removing task-discriminative information
    Assumed in Section III-C2; no ablation is reported for this component.
  • domain assumption Random frequency-cutoff augmentation in the Fourier domain improves spectral robustness without destroying task-relevant signals
    Introduced in pretraining (Section III-B-b); its benefit is not isolated in ablations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EEG-PRIME: Prototype-Aligned Representation Learning with Multi-Level Conditioning for EEG Decoding." pith.science (2026). https://pith.science/paper/DXWGJMR2

@misc{pith2026260813072,
  author       = {Pith},
  title        = {Pith review of: EEG-PRIME: Prototype-Aligned Representation Learning with Multi-Level Conditioning for EEG Decoding},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DXWGJMR2}},
  note         = {Machine review of arXiv:2608.13072}
}
read the original abstract

Electroencephalography (EEG) decoding models often generalize poorly across datasets and subjects due to domain shifts in acquisition protocols and individual neurophysiology. We propose EEG-PRIME, a two-stage EEG foundation model for cross-dataset multi-task decoding. EEG-PRIME combines masked pretraining with prototype-aligned instruction tuning to enable instruction-aware and subject-invariant decoding across diverse BCI paradigms. During pretraining, an EEG encoder learns transferable representations through masked reconstruction with frequency-cutoff spectral augmentation. During instruction tuning, EEG-PRIME incorporates task-semantic, dataset-specific, and subject-invariant conditioning. The resulting conditioning signal modulates the Q-Former through Layer-wise Query Modulation, while frozen text embeddings of class labels serve as prototypes for cosine-similarity-based prediction across heterogeneous label spaces. Experiments on sixteen datasets covering motor imagery, emotion recognition, ADHD detection, covert speech, and mental workload show consistent improvements over state-of-the-art baselines and prior EEG foundation models under cross-subject settings. On two additional held-out datasets, EEG-PRIME achieves balanced accuracy comparable to within-session calibration models without target-domain optimization, calibration, or linear probing, demonstrating promising zero-shot transfer capability.

Figures

Figures reproduced from arXiv: 2608.13072 by the authors.

Figure 1
Figure 1. Overview of EEG-PRIME and comparison with existing EEG foundation models. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The architecture design of EEG-PRIME. (a) Self-supervised pretraining of an EEG encoder with frequency-cutoff augmentation and masked reconstruction. (b) Prototype-aligned instruction tuning with multi-level conditioning, where task and dataset instruction embeddings modulate intermediate activations via Layer-wise Query Modulation (LQM) and learnable queries summarize EEG tokens through cross-attention. a) EEG data… view at source ↗
Figure 3
Figure 3. Layer-wise Query Modulation (LQM) in the [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: EEG-PRIME employs the 10–10 system with 65 [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Zero-shot inference results on two additional unseen datasets. After training, the model is evaluated directly on unseen [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Comparison of text encoders during instruction tuning. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Joint t-SNE visualization of EEG embeddings and text prototypes in the shared semantic space. Each point represents [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: ERD–attention alignment of the Q-Former. [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Analysis of the LQM mechanism in the Q-Former across all 16 EEG decoding tasks. [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: Saliency topomaps for representative tasks. Gradient [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 21 canonical work pages

  1. [5]

    Neurolm: A universal multi- task foundation model for bridging the gap between language and eeg signals,

    W.-B. Jiang, Y . Wang, B.-L. Lu, and D. Li, “Neurolm: A universal multi- task foundation model for bridging the gap between language and eeg signals,”arXiv preprint arXiv:2409.00101, 2024

    arXiv 2024

  2. [1]

    Non-invasive brain-computer interfaces: state of the art and trends,

    B. J. Edelman, S. Zhang, G. Schalk, P. Brunner, G. M ¨uller-Putz, C. Guan, and B. He, “Non-invasive brain-computer interfaces: state of the art and trends,”IEEE reviews in biomedical engineering, vol. 18, pp. 26–49, 2024

    work page 2024

  3. [2]

    Eegpt: Pretrained transformer for universal and reliable representation of eeg signals,

    G. Wang, W. Liu, Y . He, C. Xu, L. Ma, and H. Li, “Eegpt: Pretrained transformer for universal and reliable representation of eeg signals,”Ad- vances in Neural Information Processing Systems, vol. 37, pp. 39 249– 39 280, 2024

    work page 2024

  4. [3]

    Large brain model for learning generic representations with tremendous eeg data in bci,

    W.-B. Jiang, L.-M. Zhao, and B.-L. Lu, “Large brain model for learning generic representations with tremendous eeg data in bci,”arXiv preprint arXiv:2405.18765, 2024

    arXiv 2024

  5. [4]

    Cbramod: A criss-cross brain foundation model for eeg decoding,

    J. Wang, S. Zhao, Z. Luo, Y . Zhou, H. Jiang, S. Li, T. Li, and G. Pan, “Cbramod: A criss-cross brain foundation model for eeg decoding,” arXiv preprint arXiv:2412.07236, 2024

    arXiv 2024

  6. [6]

    Bert: Pre-training of deep bidirectional transformers for language understanding,

    J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “Bert: Pre-training of deep bidirectional transformers for language understanding,” inPro- ceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers), 2019, pp. 4171–4186

    2019

  7. [7]

    Language models are unsupervised multitask learners,

    A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, I. Sutskeveret al., “Language models are unsupervised multitask learners,”OpenAI blog, vol. 1, no. 8, p. 9, 2019

    2019

  8. [8]

    Eeg2rep: enhancing self-supervised eeg representation through informative masked inputs,

    N. Mohammadi Foumani, G. Mackellar, S. Ghane, S. Irtza, N. Nguyen, and M. Salehi, “Eeg2rep: enhancing self-supervised eeg representation through informative masked inputs,” inProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024, pp. 5544–5555

    work page 2024

Show all 47 references
  1. [9]

    Wavelet2vec: A filter bank masked autoencoder for eeg-based seizure subtype classification,

    R. Peng, C. Zhao, Y . Xu, J. Jiang, G. Kuang, J. Shao, and D. Wu, “Wavelet2vec: A filter bank masked autoencoder for eeg-based seizure subtype classification,” inICASSP 2023-2023 IEEE International Con- ference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2023, pp. 1–5

    2023

  2. [10]

    Biot: Biosignal transformer for cross-data learning in the wild,

    C. Yang, M. Westover, and J. Sun, “Biot: Biosignal transformer for cross-data learning in the wild,”Advances in Neural Information Pro- cessing Systems, vol. 36, pp. 78 240–78 260, 2023

    2023

  3. [11]

    Mirepnet: A pipeline and foundation model for eeg-based motor imagery classification,

    D. Liu, Z. Chen, J. Luo, S. Lian, and D. Wu, “Mirepnet: A pipeline and foundation model for eeg-based motor imagery classification,”arXiv preprint arXiv:2507.20254, 2025

    2025 arXiv

  4. [12]

    Towards robust multimodal physiological foundation models: Handling arbitrary missing modali- ties,

    W.-B. Jiang, X. Fu, Y . Ding, and C. Guan, “Towards robust multimodal physiological foundation models: Handling arbitrary missing modali- ties,”arXiv preprint arXiv:2504.19596, 2025

    2025

  5. [13]

    Sentence-bert: Sentence embeddings using siamese bert-networks,

    N. Reimers and I. Gurevych, “Sentence-bert: Sentence embeddings using siamese bert-networks,” inProceedings of the 2019 conference on empirical methods in natural language processing and the 9th international joint conference on natural language processing (EMNLP- IJCNLP), 201...

    2019

  6. [14]

    Mindfulness improves brain–computer interface performance by increasing control over neural activity in the alpha band,

    J. R. Stieger, S. Engel, H. Jiang, C. C. Cline, M. J. Kreitzer, and B. He, “Mindfulness improves brain–computer interface performance by increasing control over neural activity in the alpha band,”Cerebral Cortex, vol. 31, no. 1, pp. 426–438, 2021

    2021

  7. [15]

    Identifying similarities and differences in emotion recognition with eeg and eye movements among chinese, german, and french people,

    W. Liu, W.-L. Zheng, Z. Li, S.-Y . Wu, L. Gan, and B.-L. Lu, “Identifying similarities and differences in emotion recognition with eeg and eye movements among chinese, german, and french people,”Journal of Neural Engineering, vol. 19, no. 2, p. 026012, 2022

    2022

  8. [16]

    Investigating the effects of sleep conditions on emotion responses with eeg signals and eye movements,

    Z. Li, L.-Y . Tao, R.-X. Ma, W.-L. Zheng, and B.-L. Lu, “Investigating the effects of sleep conditions on emotion responses with eeg signals and eye movements,”IEEE Transactions on Affective Computing, 2025. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 15

    2025

  9. [17]

    Chineseeeg: A chinese linguistic corpora eeg dataset for semantic alignment and neural decoding,

    X. Mou, C. He, L. Tan, J. Yu, H. Liang, J. Zhang, Y . Tian, Y .-F. Yang, T. Xu, Q. Wanget al., “Chineseeeg: A chinese linguistic corpora eeg dataset for semantic alignment and neural decoding,”Scientific Data, vol. 11, no. 1, p. 550, 2024

    2024

  10. [18]

    Chisco: An eeg-based bci dataset for decoding of imagined speech,

    Z. Zhang, X. Ding, Y . Bao, Y . Zhao, X. Liang, B. Qin, and T. Liu, “Chisco: An eeg-based bci dataset for decoding of imagined speech,” Scientific Data, vol. 11, no. 1, p. 1265, 2024

    2024

  11. [19]

    Identification of perceived sentences using deep neural networks in eeg,

    C. Valle, C. Mendez-Orellana, C. Herff, and M. Rodriguez-Fernandez, “Identification of perceived sentences using deep neural networks in eeg,”Journal of neural engineering, vol. 21, no. 5, p. 056044, 2024

    2024

  12. [20]

    Thinking out loud, an open-access eeg-based bci dataset for inner speech recognition,

    N. Nieto, V . Peterson, H. L. Rufiner, J. E. Kamienkowski, and R. Spies, “Thinking out loud, an open-access eeg-based bci dataset for inner speech recognition,”Scientific data, vol. 9, no. 1, p. 52, 2022

    2022

  13. [21]

    Eeg dataset and openbmi toolbox for three bci paradigms: An investigation into bci illiteracy,

    M.-H. Lee, O.-Y . Kwon, Y .-J. Kim, H.-K. Kim, Y .-E. Lee, J. Williamson, S. Fazli, and S.-W. Lee, “Eeg dataset and openbmi toolbox for three bci paradigms: An investigation into bci illiteracy,”GigaScience, vol. 8, no. 5, p. giz002, 2019

    2019

  14. [22]

    Review of the bci competition iv,

    M. Tangermann, K.-R. M ¨uller, A. Aertsen, N. Birbaumer, C. Braun, C. Brunner, R. Leeb, C. Mehring, K. J. Miller, G. R. M ¨uller-Putzet al., “Review of the bci competition iv,”Frontiers in neuroscience, vol. 6, p. 55, 2012

    2012

  15. [23]

    2020 international brain–computer interface competition: A review,

    J.-H. Jeong, J.-H. Cho, Y .-E. Lee, S.-H. Lee, G.-H. Shin, Y .-S. Kweon, J. d. R. Mill ´an, K.-R. M ¨uller, and S.-W. Lee, “2020 international brain–computer interface competition: A review,”Frontiers in human neuroscience, vol. 16, p. 898300, 2022

    2020

  16. [24]

    A large eeg dataset for studying cross-session variability in motor imagery brain-computer interface,

    J. Ma, B. Yang, W. Qiu, Y . Li, S. Gao, and X. Xia, “A large eeg dataset for studying cross-session variability in motor imagery brain-computer interface,”Scientific Data, vol. 9, no. 1, p. 531, 2022

    2022

  17. [25]

    Deep learning with convolutional neural networks for eeg decoding and visualization,

    R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fiederer, M. Glasstetter, K. Eggensperger, M. Tangermann, F. Hutter, W. Burgard, and T. Ball, “Deep learning with convolutional neural networks for eeg decoding and visualization,”Human brain mapping, vol. 38, no. 11, pp. 5391–...

    2017

  18. [26]

    Eeg datasets for motor imagery brain–computer interface,

    H. Cho, M. Ahn, S. Ahn, M. Kwon, and S. C. Jun, “Eeg datasets for motor imagery brain–computer interface,”GigaScience, vol. 6, no. 7, p. gix034, 2017

    2017

  19. [27]

    Open access dataset for eeg+ nirs single-trial classification,

    J. Shin, A. von L ¨uhmann, B. Blankertz, D.-W. Kim, J. Jeong, H.-J. Hwang, and K.-R. M¨uller, “Open access dataset for eeg+ nirs single-trial classification,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 25, no. 10, pp. 1735–1745, 2016

    2016

  20. [28]

    Bci2000: a general-purpose brain-computer interface (bci) system,

    G. Schalk, D. J. McFarland, T. Hinterberger, N. Birbaumer, and J. R. Wolpaw, “Bci2000: a general-purpose brain-computer interface (bci) system,”IEEE Transactions on biomedical engineering, vol. 51, no. 6, pp. 1034–1043, 2004

    2004

  21. [29]

    Experimenters’ influ- ence on mental-imagery based brain-computer interface user training,

    L. Pillette, A. Roc, B. N’kaoua, and F. Lotte, “Experimenters’ influ- ence on mental-imagery based brain-computer interface user training,” International Journal of Human-Computer Studies, vol. 149, p. 102603, 2021

    2021

  22. [30]

    Evaluation of eeg oscillatory patterns and cognitive process during simple and compound limb motor imagery,

    W. Yi, S. Qiu, K. Wang, H. Qi, L. Zhang, P. Zhou, F. He, and D. Ming, “Evaluation of eeg oscillatory patterns and cognitive process during simple and compound limb motor imagery,”PloS one, vol. 9, no. 12, p. e114853, 2014

    2014

  23. [31]

    A large finer-grained affective computing eeg dataset,

    J. Chen, X. Wang, C. Huang, X. Hu, X. Shen, and D. Zhang, “A large finer-grained affective computing eeg dataset,”Scientific Data, vol. 10, no. 1, p. 740, 2023

    2023

  24. [32]

    Differential entropy feature for eeg-based emotion classification,

    R.-N. Duan, J.-Y . Zhu, and B.-L. Lu, “Differential entropy feature for eeg-based emotion classification,” in2013 6th international IEEE/EMBS conference on neural engineering (NER). IEEE, 2013, pp. 81–84

    2013

  25. [33]

    Comparing recognition performance and robustness of multimodal deep learning models for multimodal emotion recognition,

    W. Liu, J.-L. Qiu, W.-L. Zheng, and B.-L. Lu, “Comparing recognition performance and robustness of multimodal deep learning models for multimodal emotion recognition,”IEEE Transactions on Cognitive and Developmental Systems, vol. 14, no. 2, pp. 715–729, 2021

    2021

  26. [34]

    Seed-vii: A multi- modal dataset of six basic emotions with continuous labels for emotion recognition,

    W.-B. Jiang, X.-H. Liu, W.-L. Zheng, and B.-L. Lu, “Seed-vii: A multi- modal dataset of six basic emotions with continuous labels for emotion recognition,”IEEE Transactions on Affective Computing, vol. 16, no. 2, pp. 969–985, 2024

    2024

  27. [35]

    EEG data for ADHD / Control children,

    A. M. Nasrabadi, A. Allahverdy, M. Samavati, and M. R. Mohammadi, “EEG data for ADHD / Control children,” IEEE Dataport, 2020. [Online]. Available: https://dx.doi.org/10.21227/rzfh-zn36

    2020 doi

  28. [36]

    Electroencephalograms during mental arithmetic task performance,

    I. Zyma, S. Tukaev, I. Seleznov, K. Kiyono, A. Popov, M. Chernykh, and O. Shpenkov, “Electroencephalograms during mental arithmetic task performance,”Data, vol. 4, no. 1, p. 14, 2019

    2019

  29. [37]

    Eegnet: a compact convolutional neural network for eeg-based brain–computer interfaces,

    V . J. Lawhern, A. J. Solon, N. R. Waytowich, S. M. Gordon, C. P. Hung, and B. J. Lance, “Eegnet: a compact convolutional neural network for eeg-based brain–computer interfaces,”Journal of neural engineering, vol. 15, no. 5, p. 056013, 2018

    2018

  30. [38]

    Tsception: Capturing temporal dynamics and spatial asymmetry from eeg for emo- tion recognition,

    Y . Ding, N. Robinson, S. Zhang, Q. Zeng, and C. Guan, “Tsception: Capturing temporal dynamics and spatial asymmetry from eeg for emo- tion recognition,”IEEE Transactions on affective computing, vol. 14, no. 3, pp. 2238–2250, 2022

    2022

  31. [39]

    Transformer-based spatial-temporal feature learning for eeg decoding,

    Y . Song, X. Jia, L. Yang, and L. Xie, “Transformer-based spatial-temporal feature learning for eeg decoding,”arXiv preprint arXiv:2106.11170, 2021

    2021 arXiv

  32. [40]

    Eeg conformer: Convolutional transformer for eeg decoding and visualization,

    Y . Song, Q. Zheng, B. Liu, and X. Gao, “Eeg conformer: Convolutional transformer for eeg decoding and visualization,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 710–719, 2022

    2022

  33. [41]

    A large eeg database with users’ profile information for motor imagery brain- computer interface research,

    P. Dreyer, A. Roc, L. Pillette, S. Rimbert, and F. Lotte, “A large eeg database with users’ profile information for motor imagery brain- computer interface research,”Scientific Data, vol. 10, no. 1, p. 580, 2023

    2023

  34. [42]

    Svm-enhanced attention mechanisms for motor imagery eeg classification in brain-computer interfaces,

    Z. Otarbay and A. Kyzyrkanov, “Svm-enhanced attention mechanisms for motor imagery eeg classification in brain-computer interfaces,” Frontiers in Neuroscience, vol. 19, p. 1622847, 2025

    2025

  35. [43]

    Roberta: A robustly optimized bert pretraining approach,

    Y . Liu, M. Ott, N. Goyal, J. Du, M. Joshi, D. Chen, O. Levy, M. Lewis, L. Zettlemoyer, and V . Stoyanov, “Roberta: A robustly optimized bert pretraining approach,”arXiv preprint arXiv:1907.11692, 2019

    1907 arXiv

  36. [44]

    Deberta: Decoding-enhanced bert with disentangled attention,

    P. He, X. Liu, J. Gao, and W. Chen, “Deberta: Decoding-enhanced bert with disentangled attention,”arXiv preprint arXiv:2006.03654, 2020

    2006 arXiv

  37. [45]

    Learning eeg representations with weighted convolutional siamese network: A large multi-session post- stroke rehabilitation study,

    S. Zhang, K. K. Ang, D. Zheng, Q. Hui, X. Chen, Y . Li, N. Tang, E. Chew, R. Y . Lim, and C. Guan, “Learning eeg representations with weighted convolutional siamese network: A large multi-session post- stroke rehabilitation study,”IEEE Transactions on Neural Systems and Rehabi...

    2022

  38. [46]

    Emt: A novel transformer for generalized cross-subject eeg emotion recognition,

    Y . Ding, C. Tong, S. Zhang, M. Jiang, Y . Li, K. J. Lim, and C. Guan, “Emt: A novel transformer for generalized cross-subject eeg emotion recognition,”IEEE Transactions on Neural Networks and Learning Systems, 2025

    2025

  39. [47]

    Decoding covert speech from eeg by functional areas spatio-temporal transformer,

    M. Jiang, W. Zhang, Y . Ding, K. A. C. Teo, L. Fong, S. Zhang, Z. Guo, C. Liu, R. Bhuvanakantham, W. K. J. Simet al., “Decoding covert speech from eeg by functional areas spatio-temporal transformer,”IEEE Journal of Biomedical and Health Informatics, 2026

    2026

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.