pith. the verified trust layer for science. sign in

arxiv: 2508.18187 · v2 · submitted 2025-08-25 · 💻 cs.CV · cs.AI

BRAIN: Bias-Mitigation Continual Learning Approach to Vision-Brain Understanding

Xuan-Bac Nguyen , Thanh-Dat Truong , Pawan Sinha , Khoa Luu This is my paper

Pith reviewed 2026-05-18 21:10 UTC · model grok-4.3

classification 💻 cs.CV cs.AI
keywords continual learningbias mitigationbrain signalsvision-brain understandingcontrastive learningforgetting mitigationmemory decay
0
0 comments X p. Extension

The pith

Shifting brain signals create compounding bias that a continual learning method can correct for vision understanding

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

Brain signals weaken and shift across recording sessions because of memory decay, creating uncertain representations with poor visual context. This shift produces compounding bias that harms the training of vision-brain understanding models. The paper introduces BRAIN, a continual learning method that mitigates bias at each training step through a new de-bias contrastive loss. To keep knowledge from earlier sessions, it adds an angular-based forgetting mitigation technique. Tests show this combination delivers state-of-the-art results that beat earlier methods and those without continual learning.

Core claim

Brain signals exhibit inconsistency across recording sessions that creates growing bias in vision-brain models. The BRAIN approach addresses this by training continually and applying a de-bias contrastive learning loss to mitigate bias accumulation, combined with angular-based forgetting mitigation to retain prior knowledge without catastrophic forgetting.

What carries the argument

The Bias-Mitigation Continual Learning (BRAIN) setup with De-bias Contrastive Learning loss to reduce bias from shifting representations and Angular-based Forgetting Mitigation to preserve knowledge across sessions.

If this is right

  • The model maintains performance despite weakening brain signals over multiple sessions.
  • Vision-brain understanding benefits from sequential training without bias buildup.
  • Prior methods are outperformed on various benchmarks by handling session shifts explicitly.
  • Catastrophic forgetting is prevented while adapting to new data.

Where Pith is reading between the lines

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

  • This framework might extend to other domains with drifting data distributions such as longitudinal medical studies.
  • Testing the method on larger-scale brain datasets could reveal scalability limits.
  • Integrating it with real-time feedback systems may improve adaptive brain-computer interfaces.

Load-bearing premise

Brain signal shifts over sessions can be mitigated by the continual learning setup and de-bias loss without introducing new uncontrolled biases.

What would settle it

A test where applying the BRAIN method on held-out sessions shows no reduction in bias or performance compared to standard training.

read the original abstract

Memory decay makes it harder for the human brain to recognize visual objects and retain details. Consequently, recorded brain signals become weaker, uncertain, and contain poor visual context over time. This paper presents one of the first vision-learning approaches to address this problem. First, we statistically and experimentally demonstrate the existence of inconsistency in brain signals and its impact on the Vision-Brain Understanding (VBU) model. Our findings show that brain signal representations shift over recording sessions, leading to compounding bias, which poses challenges for model learning and degrades performance. Then, we propose a new Bias-Mitigation Continual Learning (BRAIN) approach to address these limitations. In this approach, the model is trained in a continual learning setup and mitigates the growing bias from each learning step. A new loss function named De-bias Contrastive Learning is also introduced to address the bias problem. In addition, to prevent catastrophic forgetting, where the model loses knowledge from previous sessions, the new Angular-based Forgetting Mitigation approach is introduced to preserve learned knowledge in the model. Finally, the empirical experiments demonstrate that our approach achieves State-of-the-Art (SOTA) performance across various benchmarks, surpassing prior and non-continual learning methods.

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.

Referee Report

3 major / 2 minor

Summary. The manuscript introduces BRAIN, a bias-mitigation continual learning approach for Vision-Brain Understanding (VBU). It first claims to statistically and experimentally demonstrate inconsistency and session-wise shifts in brain signals that cause compounding bias and performance degradation. It then proposes training in a continual learning setup augmented by a De-bias Contrastive Learning loss and an Angular-based Forgetting Mitigation technique to reduce bias while preventing catastrophic forgetting, ultimately reporting SOTA results across benchmarks that surpass prior and non-continual methods.

Significance. If the central empirical claims hold after proper controls and ablations, the work would address a practically relevant issue in multi-session brain-signal modeling for vision tasks. The combination of continual learning with targeted de-biasing losses targets a plausible source of distribution shift. However, the absence of verifiable experimental details, quantitative bias metrics, and isolation of the proposed mechanisms from confounding factors substantially limits the assessed significance at present.

major comments (3)
  1. [Abstract] Abstract: the assertion that brain signal representations shift over recording sessions 'leading to compounding bias' is presented without naming the quantitative measure of shift (e.g., inter-session MMD, prototype drift, or cosine distance), the statistical test employed, sample sizes, or p-values, rendering the 'statistical demonstration' unverifiable.
  2. [Experimental results] Experimental results section: the SOTA claim and the assertion that the continual setup plus De-bias Contrastive Loss and Angular Forgetting Mitigation systematically reduce bias are unsupported by any reported dataset sizes, error bars, baseline implementations, or ablation tables that hold data order fixed versus shuffled; without these, gains cannot be isolated from architecture or hyper-parameter effects.
  3. [Method] Method section: the De-bias Contrastive Learning loss is introduced to 'address the bias problem' yet no equation or pseudocode is supplied showing how it differs from standard contrastive losses or how it reduces a measurable bias metric while preserving task accuracy; the same holds for the Angular-based Forgetting Mitigation formulation.
minor comments (2)
  1. [Abstract] Abstract: the phrase 'one of the first vision-learning approaches' should be replaced by a precise literature comparison with citations.
  2. [Method] Notation: the manuscript should define all loss hyperparameters and their selection procedure explicitly rather than leaving them as free parameters.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback. We agree that additional quantitative details, experimental controls, and explicit formulations are needed to strengthen verifiability. We will revise the manuscript accordingly and address each major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the assertion that brain signal representations shift over recording sessions 'leading to compounding bias' is presented without naming the quantitative measure of shift (e.g., inter-session MMD, prototype drift, or cosine distance), the statistical test employed, sample sizes, or p-values, rendering the 'statistical demonstration' unverifiable.

    Authors: We agree that the abstract and main text should explicitly name the measures and tests. In the revision we will add the specific quantitative measures (inter-session MMD and average cosine distance between session prototypes), the statistical test (paired Wilcoxon signed-rank test), sample sizes per session, and the resulting p-values to make the demonstration of session-wise shifts fully verifiable. revision: yes

  2. Referee: [Experimental results] Experimental results section: the SOTA claim and the assertion that the continual setup plus De-bias Contrastive Loss and Angular Forgetting Mitigation systematically reduce bias are unsupported by any reported dataset sizes, error bars, baseline implementations, or ablation tables that hold data order fixed versus shuffled; without these, gains cannot be isolated from architecture or hyper-parameter effects.

    Authors: We acknowledge the need for fuller experimental reporting. The revision will include exact dataset sizes and session splits, results with mean and standard error bars over multiple random seeds, implementation details and hyperparameters for all baselines, and new ablation tables that compare the continual (fixed-order) setting against shuffled-order training to isolate the contribution of the proposed continual learning and de-biasing components. revision: yes

  3. Referee: [Method] Method section: the De-bias Contrastive Learning loss is introduced to 'address the bias problem' yet no equation or pseudocode is supplied showing how it differs from standard contrastive losses or how it reduces a measurable bias metric while preserving task accuracy; the same holds for the Angular-based Forgetting Mitigation formulation.

    Authors: We will expand the method section with the complete equations for both losses and accompanying pseudocode. The De-bias Contrastive Loss will be shown as a modified InfoNCE objective that incorporates session-aware negative reweighting to reduce a defined bias metric (session-wise representation drift). The Angular-based Forgetting Mitigation will be presented as an angular-distance regularizer on the feature manifold. We will also include a short analysis demonstrating that these terms reduce the bias metric without degrading task accuracy. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical demonstration and novel losses stand independently

full rationale

The paper first claims to statistically and experimentally demonstrate session-wise shifts in brain signals and resulting compounding bias, then introduces a continual learning setup together with two explicitly new components (De-bias Contrastive Learning loss and Angular-based Forgetting Mitigation) whose purpose is to counteract those shifts. The SOTA performance is asserted solely on the basis of benchmark experiments rather than any equation that equates a fitted parameter to a predicted quantity or that reduces the bias-mitigation effect to a self-referential definition. No load-bearing step invokes a prior result by the same authors as an unverified uniqueness theorem, and the abstract contains no ansatz smuggled via citation or renaming of a known pattern. The derivation chain is therefore self-contained against external benchmarks and does not collapse to its own inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption of measurable signal shift and bias accumulation, plus the effectiveness of the introduced loss and mitigation techniques; no new physical entities or free parameters are explicitly listed beyond standard ML hyperparameters.

free parameters (1)
  • loss function hyperparameters
    The de-bias contrastive loss and angular mitigation likely involve tunable weights or margins fitted during training.
axioms (1)
  • domain assumption Brain signal representations shift over recording sessions leading to compounding bias
    Invoked in the abstract as the foundation for the bias-mitigation approach.

pith-pipeline@v0.9.0 · 5756 in / 1287 out tokens · 18056 ms · 2026-05-18T21:10:01.507084+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

130 extracted references · 130 canonical work pages · 13 internal anchors

  1. [1]

    Science 255(5043), 419–423 (1992)

    Van Essen, D.C., Anderson, C.H., Felleman, D.J.: Information processing in the primate visual system: an integrated systems perspective. Science 255(5043), 419–423 (1992)

    work page 1992

  2. [2]

    Science 311(5761), 670–674 (2006)

    Tsao, D.Y., Freiwald, W.A., Tootell, R.B., Livingstone, M.S.: A cortical region consisting entirely of face-selective cells. Science 311(5761), 670–674 (2006)

    work page 2006

  3. [3]

    Cell 173(6), 1343–1355 (2018)

    Liang, L., Fratzl, A., Goldey, G., Ramesh, R.N., Sugden, A.U., Morgan, J.L., Chen, C., Andermann, M.L.: A fine-scale functional logic to convergence from retina to thalamus. Cell 173(6), 1343–1355 (2018)

    work page 2018

  4. [4]

    Science 272(5268), 1665–1668 (1996)

    Wang, G., Tanaka, K., Tanifuji, M.: Optical imaging of functional organization in the monkey inferotemporal cortex. Science 272(5268), 1665–1668 (1996)

    work page 1996

  5. [5]

    Proceedings of the national academy of sciences 98(2), 676–682 (2001)

    Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A., Shulman, G.L.: A default mode of brain function. Proceedings of the national academy of sciences 98(2), 676–682 (2001)

    work page 2001

  6. [6]

    J physiol 148(3), 574–591 (1959)

    Hubel, D.H., Wiesel, T.N., et al.: Receptive fields of single neurones in the cat’s striate cortex. J physiol 148(3), 574–591 (1959)

    work page 1959

  7. [7]

    Nature neuroscience 15(12), 1683–1690 (2012)

    Nauhaus, I., Nielsen, K.J., Disney, A.A., Callaway, E.M.: Orthogonal micro- organization of orientation and spatial frequency in primate primary visual cortex. Nature neuroscience 15(12), 1683–1690 (2012)

    work page 2012

  8. [8]

    Journal of Neuroscience 4(1), 309–356 (1984)

    Livingstone, M.S., Hubel, D.H.: Anatomy and physiology of a color system in the primate visual cortex. Journal of Neuroscience 4(1), 309–356 (1984)

    work page 1984

  9. [9]

    Journal of experimental psychology: General 109(2), 160 (1980)

    Posner, M.I., Snyder, C.R., Davidson, B.J.: Attention and the detection of signals. Journal of experimental psychology: General 109(2), 160 (1980)

    work page 1980

  10. [10]

    The Journal of physiology 195(1), 215–243 (1968)

    Hubel, D.H., Wiesel, T.N.: Receptive fields and functional architecture of monkey striate cortex. The Journal of physiology 195(1), 215–243 (1968)

    work page 1968

  11. [11]

    : A massive 7t fmri dataset to bridge cognitive neuroscience and artificial intelligence

    Allen, E.J., St-Yves, G., Wu, Y., Breedlove, J.L., Prince, J.S., Dowdle, L.T., Nau, M., Caron, B., Pestilli, F., Charest, I., et al. : A massive 7t fmri dataset to bridge cognitive neuroscience and artificial intelligence. Nature neuroscience 25(1), 116–126 (2022)

    work page 2022

  12. [12]

    Elife 12, 82580 (2023)

    Hebart, M.N., Contier, O., Teichmann, L., Rockter, A.H., Zheng, C.Y., Kidder, A., Corriveau, A., Vaziri-Pashkam, M., Baker, C.I.: Things-data, a multimodal collection of large-scale datasets for investigating object representations in human brain and behavior. Elife 12, 82580 (2023)

    work page 2023

  13. [13]

    In: 26th Annual Meeting of the Organization for Human Brain Mapping (2020)

    Boyle, J.A., Pinsard, B., Boukhdhir, A., Belleville, S., Brambatti, S., Chen, J., Cohen-Adad, J., Cyr, A., Fuente Rainville, P., Bellec, P.: The courtois 20 project on neuronal modelling-first data release. In: 26th Annual Meeting of the Organization for Human Brain Mapping (2020)

    work page 2020

  14. [14]

    Trends in cogni- tive sciences 15(12), 567–575 (2011)

    Block, N.: Perceptual consciousness overflows cognitive access. Trends in cogni- tive sciences 15(12), 567–575 (2011)

    work page 2011

  15. [15]

    Proceedings of the National Academy of Sciences 108(27), 11252–11255 (2011)

    Buschman, T.J., Siegel, M., Roy, J.E., Miller, E.K.: Neural substrates of cog- nitive capacity limitations. Proceedings of the National Academy of Sciences 108(27), 11252–11255 (2011)

    work page 2011

  16. [16]

    Cohen, M.A., Dennett, D.C., Kanwisher, N.: What is the bandwidth of perceptual experience? Trends in cognitive sciences 20(5), 324–335 (2016)

    work page 2016

  17. [17]

    Behavioral and brain sciences 22(3), 341–365 (1999)

    Pylyshyn, Z.: Is vision continuous with cognition?: The case for cognitive impen- etrability of visual perception. Behavioral and brain sciences 22(3), 341–365 (1999)

    work page 1999

  18. [18]

    Annual review of psychology 69(1), 105–129 (2018)

    Whitney, D., Yamanashi Leib, A.: Ensemble perception. Annual review of psychology 69(1), 105–129 (2018)

    work page 2018

  19. [19]

    Trends in cognitive sciences 17(8), 391–400 (2013)

    Luck, S.J., Vogel, E.K.: Visual working memory capacity: from psychophysics and neurobiology to individual differences. Trends in cognitive sciences 17(8), 391–400 (2013)

    work page 2013

  20. [20]

    In: 2022 International Joint Conference on Neural Networks (IJCNN), pp

    Ozcelik, F., Choksi, B., Mozafari, M., Reddy, L., VanRullen, R.: Reconstruction of perceived images from fmri patterns and semantic brain exploration using instance-conditioned gans. In: 2022 International Joint Conference on Neural Networks (IJCNN), pp. 1–8 (2022). IEEE

    work page 2022

  21. [21]

    PLoS computational biology 15(1), 1006633 (2019)

    Shen, G., Horikawa, T., Majima, K., Kamitani, Y.: Deep image reconstruction from human brain activity. PLoS computational biology 15(1), 1006633 (2019)

    work page 2019

  22. [22]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Takagi, Y., Nishimoto, S.: High-resolution image reconstruction with latent dif- fusion models from human brain activity. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 14453–14463 (2023)

    work page 2023

  23. [23]

    Advances in Neural Information Processing Systems 36 (2024)

    Scotti, P., Banerjee, A., Goode, J., Shabalin, S., Nguyen, A., Dempster, A., Verlinde, N., Yundler, E., Weisberg, D., Norman, K., et al.: Reconstructing the mind’s eye: fmri-to-image with contrastive learning and diffusion priors. Advances in Neural Information Processing Systems 36 (2024)

    work page 2024

  24. [24]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Wang, S., Liu, S., Tan, Z., Wang, X.: Mindbridge: A cross-subject brain decoding framework. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11333–11342 (2024)

    work page 2024

  25. [25]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Quan, R., Wang, W., Tian, Z., Ma, F., Yang, Y.: Psychometry: An omnifit 21 model for image reconstruction from human brain activity. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 233–243 (2024)

    work page 2024

  26. [26]

    arXiv preprint arXiv:2403.18211 (2024)

    Huo, J., Wang, Y., Qian, X., Wang, Y., Li, C., Feng, J., Fu, Y.: Neuropic- tor: Refining fmri-to-image reconstruction via multi-individual pretraining and multi-level modulation. arXiv preprint arXiv:2403.18211 (2024)

    work page arXiv 2024

  27. [27]

    arXiv preprint arXiv:2403.11207 (2024)

    Scotti, P.S., Tripathy, M., Villanueva, C.K.T., Kneeland, R., Chen, T., Narang, A., Santhirasegaran, C., Xu, J., Naselaris, T., Norman, K.A., et al.: Mindeye2: Shared-subject models enable fmri-to-image with 1 hour of data. arXiv preprint arXiv:2403.11207 (2024)

    work page arXiv 2024

  28. [28]

    https://arxiv.org/abs/2404.07202

    Xia, W., Charette, R., ¨Oztireli, C., Xue, J.-H.: UMBRAE: Unified Multimodal Brain Decoding (2024). https://arxiv.org/abs/2404.07202

    work page arXiv 2024

  29. [29]

    : Decoding motor imagery from the posterior parietal cortex of a tetraplegic human

    Aflalo, T., Kellis, S., Klaes, C., Lee, B., Shi, Y., Pejsa, K., Shanfield, K., Hayes- Jackson, S., Aisen, M., Heck, C., et al. : Decoding motor imagery from the posterior parietal cortex of a tetraplegic human. Science 348(6237), 906–910 (2015)

    work page 2015

  30. [30]

    arXiv preprint arXiv:2306.16934 (2023)

    Bai, Y., Wang, X., Cao, Y.-p., Ge, Y., Yuan, C., Shan, Y.: Dreamdiffu- sion: Generating high-quality images from brain eeg signals. arXiv preprint arXiv:2306.16934 (2023)

    work page arXiv 2023

  31. [31]

    In: Proceedings of the Ieee/cvf Conference on Computer Vision and Pattern Recognition, pp

    Nguyen, X.-B., Duong, C.N., Li, X., Gauch, S., Seo, H.-S., Luu, K.: Micron-bert: Bert-based facial micro-expression recognition. In: Proceedings of the Ieee/cvf Conference on Computer Vision and Pattern Recognition, pp. 1482–1492 (2023)

    work page 2023

  32. [32]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Nguyen, X.-B., Bui, D.T., Duong, C.N., Bui, T.D., Luu, K.: Clusformer: A transformer based clustering approach to unsupervised large-scale face and visual landmark recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10847–10856 (2021)

    work page 2021

  33. [33]

    IEEE Access 8, 162973–162981 (2020)

    Nguyen, X.-B., Lee, G.S., Kim, S.H., Yang, H.J.: Self-supervised learning based on spatial awareness for medical image analysis. IEEE Access 8, 162973–162981 (2020)

    work page 2020

  34. [34]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Nguyen, P., Quach, K.G., Duong, C.N., Le, N., Nguyen, X.-B., Luu, K.: Multi- camera multiple 3d object tracking on the move for autonomous vehicles. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2569–2578 (2022)

    work page 2022

  35. [35]

    In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pp

    Nguyen, H.-Q., Truong, T.-D., Nguyen, X.B., Dowling, A., Li, X., Luu, K.: Insect-foundation: A foundation model and large-scale 1m dataset for visual insect understanding. In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pp. 21945–21955 (2024) 22

    work page 2024

  36. [36]

    arXiv preprint arXiv:2405.19722 (2024)

    Nguyen, X.-B., Nguyen, H.-Q., Chen, S.Y.-C., Khan, S.U., Churchill, H., Luu, K.: Qclusformer: A quantum transformer-based framework for unsupervised visual clustering. arXiv preprint arXiv:2405.19722 (2024)

    work page arXiv 2024

  37. [37]

    In: 2022 26th International Conference on Pattern Recognition (ICPR), pp

    Truong, T.-D., Chappa, R.T.N., Nguyen, X.-B., Le, N., Dowling, A.P., Luu, K.: Otadapt: Optimal transport-based approach for unsupervised domain adapta- tion. In: 2022 26th International Conference on Pattern Recognition (ICPR), pp. 2850–2856 (2022). IEEE

    work page 2022

  38. [38]

    arXiv preprint arXiv:2406.00843 (2024)

    Nguyen, H.-Q., Nguyen, X.B., Chen, S.Y.-C., Churchill, H., Borys, N., Khan, S.U., Luu, K.: Diffusion-inspired quantum noise mitigation in parameterized quantum circuits. arXiv preprint arXiv:2406.00843 (2024)

    work page arXiv 2024

  39. [39]

    arXiv preprint arXiv:2309.09907 (2023)

    Nguyen, X.B., Churchill, H., Luu, K., Khan, S.U.: Quantum vision clustering. arXiv preprint arXiv:2309.09907 (2023)

    work page arXiv 2023

  40. [40]

    arXiv preprint arXiv:2205.15948 (2022)

    Nguyen, X.B., Bisht, A., Churchill, H., Luu, K.: Two-dimensional quantum material identification via self-attention and soft-labeling in deep learning. arXiv preprint arXiv:2205.15948 (2022)

    work page arXiv 2022

  41. [41]

    Quantum Machine Intelligence6(2), 61 (2024)

    Nguyen, X.-B., Nguyen, H.-Q., Churchill, H., Khan, S.U., Luu, K.: Quantum visual feature encoding revisited. Quantum Machine Intelligence6(2), 61 (2024)

    work page 2024

  42. [42]

    arXiv preprint arXiv:2304.07408 (2023)

    Nguyen, X.-B., Duong, C.N., Savvides, M., Roy, K., Churchill, H., Luu, K.: Fairness in visual clustering: A novel transformer clustering approach. arXiv preprint arXiv:2304.07408 (2023)

    work page arXiv 2023

  43. [43]

    arXiv preprint arXiv:2312.00236 (2023)

    Nguyen, X.-B., Li, X., Khan, S.U., Luu, K.: Brainformer: Modeling mri brain functions to machine vision. arXiv preprint arXiv:2312.00236 (2023)

    work page arXiv 2023

  44. [44]

    International Journal of Contents 15(4), 8–15 (2019)

    Nguyen-Xuan, B., Lee, G.-S.: Sketch recognition using lstm with attention mech- anism and minimum cost flow algorithm. International Journal of Contents 15(4), 8–15 (2019)

    work page 2019

  45. [45]

    arXiv preprint arXiv:2505.15755 (2025)

    Xia, W., Oztireli, C.: Exploring the visual feature space for multimodal neural decoding. arXiv preprint arXiv:2505.15755 (2025)

    work page arXiv 2025

  46. [46]

    arXiv preprint arXiv:2308.00262 (2023)

    Nguyen, X.-B., Liu, X., Li, X., Luu, K.: The algonauts project 2023 challenge: Uark-ualbany team solution. arXiv preprint arXiv:2308.00262 (2023)

    work page arXiv 2023

  47. [47]

    In: 2019 IEEE/CVF Inter- national Conference on Computer Vision Workshop (ICCVW), pp

    Nguyen, X.-B., Lee, G.-S., Kim, S.-H., Yang, H.-J.: Audio-video based emotion recognition using minimum cost flow algorithm. In: 2019 IEEE/CVF Inter- national Conference on Computer Vision Workshop (ICCVW), pp. 3737–3741 (2019). IEEE

    work page 2019

  48. [48]

    arXiv preprint 23 arXiv:2408.03596 (2024)

    Nguyen, X.-B., Nguyen, H.-Q., Churchill, H., Khan, S.U., Luu, K.: Hierarchi- cal quantum control gates for functional mri understanding. arXiv preprint 23 arXiv:2408.03596 (2024)

    work page arXiv 2024

  49. [49]

    International Journal of Computer Vision, 1–26 (2025)

    Truong, T.-D., Nguyen, H.-Q., Nguyen, X.-B., Dowling, A., Li, X., Luu, K.: Insect-foundation: A foundation model and large multimodal dataset for vision- language insect understanding. International Journal of Computer Vision, 1–26 (2025)

    work page 2025

  50. [50]

    arXiv preprint arXiv:2411.13378 (2024)

    Nguyen, H.-Q., Nguyen, X.-B., Churchill, H., Choudhary, A.K., Sinha, P., Khan, S.U., Luu, K.: Quantum-brain: Quantum-inspired neural network approach to vision-brain understanding. arXiv preprint arXiv:2411.13378 (2024)

    work page arXiv 2024

  51. [51]

    arXiv preprint arXiv:2405.18808 (2024)

    Nguyen, X.-B., Jang, H., Li, X., Khan, S.U., Sinha, P., Luu, K.: Bractive: A brain activation approach to human visual brain learning. arXiv preprint arXiv:2405.18808 (2024)

    work page arXiv 2024

  52. [52]

    arXiv preprint arXiv:2411.17475 (2024)

    Nguyen, X.-B., Choudhary, A.K., Sinha, P., Li, X., Luu, K.: Cobra: A continual learning approach to vision-brain understanding. arXiv preprint arXiv:2411.17475 (2024)

    work page arXiv 2024

  53. [53]

    Advances in Neural Information Processing Systems 32 (2019)

    Beliy, R., Gaziv, G., Hoogi, A., Strappini, F., Golan, T., Irani, M.: From voxels to pixels and back: Self-supervision in natural-image reconstruction from fmri. Advances in Neural Information Processing Systems 32 (2019)

    work page 2019

  54. [54]

    Advances in Neural Information Processing Systems 36, 15096–15107 (2023)

    Fang, T., Zheng, Q., Pan, G.: Alleviating the semantic gap for generalized fmri- to-image reconstruction. Advances in Neural Information Processing Systems 36, 15096–15107 (2023)

    work page 2023

  55. [55]

    arXiv preprint arXiv:2408.06788 (2024)

    Chen, H., He, L., Liu, Y., Yang, L.: Visual neural decoding via improved visual- eeg semantic consistency. arXiv preprint arXiv:2408.06788 (2024)

    work page arXiv 2024

  56. [56]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 45(9), 10760–10777 (2023)

    Du, C., Fu, K., Li, J., He, H.: Decoding visual neural representations by multi- modal learning of brain-visual-linguistic features. IEEE Transactions on Pattern Analysis and Machine Intelligence 45(9), 10760–10777 (2023)

    work page 2023

  57. [57]

    arXiv preprint arXiv:2403.07721 (2024)

    Li, D., Wei, C., Li, S., Zou, J., Qin, H., Liu, Q.: Visual decoding and reconstruc- tion via eeg embeddings with guided diffusion. arXiv preprint arXiv:2403.07721 (2024)

    work page arXiv 2024

  58. [58]

    arXiv preprint arXiv:2308.13234 (2023)

    Song, Y., Liu, B., Li, X., Shi, N., Wang, Y., Gao, X.: Decoding natural images from eeg for object recognition. arXiv preprint arXiv:2308.13234 (2023)

    work page arXiv 2023

  59. [59]

    In: International Conference on Machine Learning, pp

    Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al.: Learning transferable visual mod- els from natural language supervision. In: International Conference on Machine Learning, pp. 8748–8763 (2021). PmLR

    work page 2021

  60. [60]

    Deep Unsupervised Learning using Nonequilibrium Thermodynamics

    Sohl-Dickstein, J., Weiss, E.A., Maheswaranathan, N., Ganguli, S.: Deep 24 unsupervised learning using nonequilibrium thermodynamics. CoRR abs/1503.03585 (2015) 1503.03585

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  61. [61]

    Denoising Diffusion Probabilistic Models

    Ho, J., Jain, A., Abbeel, P.: Denoising diffusion probabilistic models. arXiv preprint arxiv:2006.11239 (2020)

    work page internal anchor Pith review Pith/arXiv arXiv 2006

  62. [62]

    Diffusion Models Beat GANs on Image Synthesis

    Dhariwal, P., Nichol, A.: Diffusion models beat gans on image synthesis. CoRR abs/2105.05233 (2021) 2105.05233

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  63. [63]

    High-Resolution Image Synthesis with Latent Diffusion Models

    Rombach, R., Blattmann, A., Lorenz, D., Esser, P., Ommer, B.: High-Resolution Image Synthesis with Latent Diffusion Models (2022). https://arxiv.org/abs/ 2112.10752

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  64. [64]

    https://arxiv.org/abs/2211

    Xu, X., Wang, Z., Zhang, E., Wang, K., Shi, H.: Versatile Diffusion: Text, Images and Variations All in One Diffusion Model (2024). https://arxiv.org/abs/2211. 08332

    work page 2024

  65. [65]

    https: //arxiv.org/abs/2112.08654

    Wang, Z., Zhang, Z., Lee, C.-Y., Zhang, H., Sun, R., Ren, X., Su, G., Perot, V., Dy, J., Pfister, T.: Learning to Prompt for Continual Learning (2022). https: //arxiv.org/abs/2112.08654

    work page arXiv 2022

  66. [66]

    Memory Efficient Experience Replay for Streaming Learning

    Hayes, T.L., Cahill, N.D., Kanan, C.: Memory Efficient Experience Replay for Streaming Learning (2019). https://arxiv.org/abs/1809.05922

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  67. [67]

    On Tiny Episodic Memories in Continual Learning

    Chaudhry, A., Rohrbach, M., Elhoseiny, M., Ajanthan, T., Dokania, P.K., Torr, P.H.S., Ranzato, M.: On Tiny Episodic Memories in Continual Learning (2019). https://arxiv.org/abs/1902.10486

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  68. [68]

    https://arxiv

    Chaudhry, A., Gordo, A., Dokania, P.K., Torr, P., Lopez-Paz, D.: Using Hind- sight to Anchor Past Knowledge in Continual Learning (2021). https://arxiv. org/abs/2002.08165

    work page arXiv 2021

  69. [69]

    Efficient Lifelong Learning with A-GEM

    Chaudhry, A., Ranzato, M., Rohrbach, M., Elhoseiny, M.: Efficient Lifelong Learning with A-GEM (2019). https://arxiv.org/abs/1812.00420

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  70. [70]

    https: //arxiv.org/abs/2004.07211

    Buzzega, P., Boschini, M., Porrello, A., Abati, D., Calderara, S.: Dark Experi- ence for General Continual Learning: a Strong, Simple Baseline (2020). https: //arxiv.org/abs/2004.07211

    work page arXiv 2020

  71. [71]

    https://arxiv.org/abs/1611

    Rebuffi, S.-A., Kolesnikov, A., Sperl, G., Lampert, C.H.: iCaRL: Incremen- tal Classifier and Representation Learning (2017). https://arxiv.org/abs/1611. 07725

    work page 2017

  72. [72]

    In: 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Aller- ton), pp

    Shokri, R., Shmatikov, V.: Privacy-preserving deep learning. In: 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Aller- ton), pp. 909–910 (2015). https://doi.org/10.1109/ALLERTON.2015.7447103

    work page doi:10.1109/allerton.2015.7447103 2015

  73. [73]

    25 https://arxiv.org/abs/2110.00175

    Pham, Q., Liu, C., Hoi, S.: DualNet: Continual Learning, Fast and Slow (2021). 25 https://arxiv.org/abs/2110.00175

    work page arXiv 2021

  74. [74]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Cha, H., Lee, J., Shin, J.: Co2l: Contrastive continual learning. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 9516–9525 (2021)

    work page 2021

  75. [75]

    Large Scale Incremental Learning

    Wu, Y., Chen, Y., Wang, L., Ye, Y., Liu, Z., Guo, Y., Fu, Y.: Large Scale Incremental Learning (2019). https://arxiv.org/abs/1905.13260

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  76. [76]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Wang, K., Herranz, L., Weijer, J.: Continual learning in cross-modal retrieval. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3628–3638 (2021)

    work page 2021

  77. [77]

    In: The Thirteenth International Conference on Learning Representations (2025)

    Liu, W., Zhu, F., Wei, L., Tian, Q.: C-clip: Multimodal continual learning for vision-language model. In: The Thirteenth International Conference on Learning Representations (2025)

    work page 2025

  78. [78]

    In: Proceed- ings of the 32nd ACM International Conference on Information and Knowledge Management, pp

    Cai, Y., Bi, K., Fan, Y., Guo, J., Chen, W., Cheng, X.: L2r: Lifelong learning for first-stage retrieval with backward-compatible representations. In: Proceed- ings of the 32nd ACM International Conference on Information and Knowledge Management, pp. 183–192 (2023)

    work page 2023

  79. [79]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Wan, T.S., Chen, J.-C., Wu, T.-Y., Chen, C.-S.: Continual learning for visual search with backward consistent feature embedding. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16702–16711 (2022)

    work page 2022

  80. [80]

    In: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Cui, Z., Zhou, J., Wang, X., Zhu, M., Peng, Y.: Learning continual compatible representation for re-indexing free lifelong person re-identification. In: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16614–16623 (2024)

    work page 2024

Showing first 80 references.