Pith. sign in

REVIEW 3 major objections 4 minor 3 cited by

Dynamic Uncertainty-aware Multimodal Fusion for Outdoor Health Monitoring

T0 review · 3 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read DUAL-Health claims uncertainty-weighted sensor fusion improves outdoor health monitoring.

desk verdict The abstract describes a plausible uncertainty-aware fusion method for MLLM-based health monitoring, but the submitted full text is a different paper, so none of the claims can be checked. read the letter →

arxiv 2508.09085 v1 pith:Y47IQFEO submitted 2025-08-12 cs.NI cs.AIcs.LG

Zihan Fang , Zheng Lin , Senkang Hu , Yihang Tao , Yiqin Deng , Xianhao Chen , Yuguang Fang This is my paper
classification cs.NIcs.AIcs.LG
keywords uncertainty-awarefusionmultimodallargelanguagemodelshealthmonitoringsensornoisemodalitysmall-datafine-tuningoutdoor
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 argues that the main obstacle to using multimodal large language models for outdoor health monitoring is not model capacity but how to handle sensor modalities with different noise levels. It introduces DUAL-Health, which estimates each modality's uncertainty from current readings and their temporal context, then assigns fusion weights accordingly and aligns modality distributions in a common semantic space. The claim is that this uncertainty-weighted fusion outperforms state-of-the-art baselines in detection accuracy and robustness under dynamic, noisy outdoor conditions. If right, it shows a practical route to reliable small-data health monitoring with pre-trained multimodal models.

What carries the argument

The uncertainty estimator that combines current features (capturing acquisition noise) with temporal features (capturing fluctuation noise) to produce a calibrated per-modality uncertainty score. This score drives the fusion-weight assignment for each modality and also guides the distribution-alignment step; it is the component that turns noisy, heterogeneous sensor data into weighted evidence.

What would settle it

Take a fixed multimodal health dataset, add controlled synthetic noise at a known level to one modality (e.g., accelerometer), and compare DUAL-Health's assigned fusion weight with the true optimal weight. If the assigned weight does not track the known noise level, the uncertainty estimator is not calibrated and the framework's core mechanism fails.

Watch

Extended reading notes

Core claim

DUAL-Health's central claim is that explicitly quantifying modality-level uncertainty—separating input noise from sensor acquisition from fluctuation noise caused by sudden physiological changes—and using that quantification to set per-modality fusion weights makes multimodal fusion robust when some modalities are noisy or missing. The framework also aligns modality distributions in a common semantic space so that recovery of missing data from fluctuating distributions becomes more accurate. On the paper's experiments, this yields higher detection accuracy and robustness than state-of-the-art baselines.

Load-bearing premise

The framework's gains rest on the assumption that modality uncertainty from input noise and fluctuation noise can be accurately quantified from current and temporal features; if that estimate is miscalibrated, the fusion weights would be wrong and the claimed robustness gains would shrink.

Editorial extensions

If this is right

  • Multimodal large language model based health monitors can be fine-tuned with small datasets rather than trained from scratch.
  • The system can degrade gracefully when one sensor becomes noisy: the uncertain modality is automatically down-weighted.
  • Missing data from a fluctuating modality can be recovered by aligning its distribution with the shared semantic space.
  • Detection accuracy and robustness in outdoor settings improve over fixed-weight fusion baselines.

Reading between the lines

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

  • If the uncertainty estimator is well calibrated, the same weighting scheme could transfer to other small-data multimodal settings such as industrial wearables or animal tracking, though the paper does not test this.
  • A direct test of the mechanism would compare DUAL-Health against an oracle that receives the true noise levels; the gap would reveal how much of the gain comes from the estimator itself versus the fusion design.
  • The temporal-feature component suggests that longer observation histories could further improve calibration; an ablation varying sequence length would make this prediction testable.
  • A limitation implied by the abstract is that the claim of accurate uncertainty quantification is asserted without derivation; synthetic-noise experiments with known ground-truth noise levels would make that assumption directly measurable.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The submission, as represented by its abstract, claims a new uncertainty-aware multimodal fusion framework called DUAL-Health for outdoor health monitoring. The framework is said to (i) quantify modality uncertainty arising from input and fluctuation noise using current and temporal features, (ii) customize fusion weights from calibrated uncertainty estimates, and (iii) align modality distributions in a common semantic space to improve missing-data recovery. The abstract further claims that extensive experiments show DUAL-Health outperforms state-of-the-art baselines in detection accuracy and robustness. However, the full text supplied with this submission is not the DUAL-Health paper: it is a multilayer network analysis of economics journals (arXiv:2508.09079v2), containing no equations, algorithms, datasets, or experiments related to multimodal health monitoring. Consequently, the central claims of the abstract are entirely unsupported by the submitted manuscript.

Significance. If the claimed results are correct, DUAL-Health would address a real and timely problem: enabling MLLM-based health monitoring under sensor noise and data scarcity. The abstract presents a coherent three-part design and a plausible motivation. However, the submitted manuscript provides no technical content for the framework and no empirical evidence. There is no derivation, no dataset description, no baseline comparison, no code, and no machine-checked verification. The significance of the work therefore cannot be assessed beyond the abstract's programmatic statement. The paper's only demonstrable strength is a clear problem framing in the abstract; none of the claimed contributions are visible in the provided full text.

major comments (3)
  1. [Full Text (entire manuscript)] The full text of the submission is a different paper: an economics journal multilayer network analysis. It contains no description of the DUAL-Health framework, no uncertainty quantification model, no fusion mechanism, no missing-data recovery method, and no health-monitoring experiments. The central claim of the abstract—that DUAL-Health outperforms state-of-the-art baselines—is therefore unsupported by the submitted manuscript. This is a load-bearing issue that prevents any evaluation of the paper's core contribution.
  2. [Abstract, first contribution] The load-bearing premise 'we accurately quantify modality uncertainty caused by input and fluctuation noise with current and temporal features' is stated without a formal definition, derivation, or validation. Because the fusion weights are customized from these uncertainty estimates, the claimed robustness gains depend entirely on the calibration of this quantification. The submitted text provides no equation, experiment, or ablation that tests this premise.
  3. [Abstract, last sentence] The claim 'Extensive experiments demonstrate that our DUAL-Health outperforms state-of-the-art baselines in detection accuracy and robustness' is not supported by any experimental material in the submission. There is no dataset, no list of sensor modalities, no baseline names, no evaluation metrics, no error bars, no ablations, and no code. For an empirical performance claim, this absence is decisive: the result cannot be checked or reproduced.
minor comments (4)
  1. [Abstract] Typo: 'muitimodal' should be 'multimodal'.
  2. [Abstract] Enumeration is inconsistent: the first challenge is marked 'I)' while the second and third use lowercase 'ii)' and 'iii)'.
  3. [Abstract] The terms 'input noise', 'fluctuation noise', 'current and temporal features' are not defined. Definitions are needed to make the proposed uncertainty quantification concrete.
  4. [Abstract] The phrase 'static multimodal deep learning frameworks... requires extensive data training from scratch' is a strong claim that needs a citation or qualification, and the grammar ('requires') should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

Cannot assess circularity: the submitted full text is an unrelated economics paper, so the DUAL-Health derivation is absent.

full rationale

The abstract describes a DUAL-Health multimodal fusion framework and claims that it outperforms state-of-the-art baselines, but the full text supplied is a completely different manuscript: 'Exploring the Shape of Economics: A Multilayer Network Analysis of Social Communities and Intellectual Similarity among Journals before and after the 2008 Financial Crisis' (arXiv:2508.09079v2). The economics paper contains no uncertainty quantification model, no fusion-weight formula, no experimental protocol, and no derivation of the claimed detection-accuracy or robustness results. Therefore, there is no derivation chain to walk, and no specific equation, fitted parameter, or self-citation relation can be exhibited that reduces the claimed prediction to its inputs. Under the hard rule that circularity may only be claimed when the paper's own text permits quoting the specific reduction, no circular step can be identified. The absence of the relevant manuscript means the central claim is unverifiable from this material, but that is a completeness problem, not evidence of circularity. The score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The abstract introduces DUAL-Health as a framework, not a new physical entity. The axioms listed are the domain assumptions implicit in the abstract's three design choices. No numeric free parameters are identifiable from the abstract alone.

assumptions (3)
  • domain assumption Sensor noise in outdoor health monitoring can be decomposed into input noise and fluctuation noise, and both can be quantified from current and temporal features.
    The entire uncertainty weighting scheme rests on the ability to measure these noise types from available features; the abstract asserts this without showing evidence.
  • domain assumption Pre-trained multimodal large language models can be effectively fine-tuned on small datasets for health monitoring.
    This is the premise for using MLLMs over conventional static deep learning; the abstract asserts it as motivation.
  • domain assumption Aligning modality distributions in a common semantic space improves recovery of missing data from fluctuating distributions.
    The proposed data-recovery step assumes a shared semantic space exists and that distribution alignment in that space helps; the abstract provides no formal support.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dynamic Uncertainty-aware Multimodal Fusion for Outdoor Health Monitoring." pith.science (2026). https://pith.science/paper/Y47IQFEO

@misc{pith2026250809085,
  author       = {Pith},
  title        = {Pith review of: Dynamic Uncertainty-aware Multimodal Fusion for Outdoor Health Monitoring},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y47IQFEO}},
  note         = {Machine review of arXiv:2508.09085}
}
read the original abstract

Outdoor health monitoring is essential to detect early abnormal health status for safeguarding human health and safety. Conventional outdoor monitoring relies on static multimodal deep learning frameworks, which requires extensive data training from scratch and fails to capture subtle health status changes. Multimodal large language models (MLLMs) emerge as a promising alternative, utilizing only small datasets to fine-tune pre-trained information-rich models for enabling powerful health status monitoring. Unfortunately, MLLM-based outdoor health monitoring also faces significant challenges: I) sensor data contains input noise stemming from sensor data acquisition and fluctuation noise caused by sudden changes in physiological signals due to dynamic outdoor environments, thus degrading the training performance; ii) current transformer based MLLMs struggle to achieve robust multimodal fusion, as they lack a design for fusing the noisy modality; iii) modalities with varying noise levels hinder accurate recovery of missing data from fluctuating distributions. To combat these challenges, we propose an uncertainty-aware multimodal fusion framework, named DUAL-Health, for outdoor health monitoring in dynamic and noisy environments. First, to assess the impact of noise, we accurately quantify modality uncertainty caused by input and fluctuation noise with current and temporal features. Second, to empower efficient muitimodal fusion with low-quality modalities,we customize the fusion weight for each modality based on quantified and calibrated uncertainty. Third, to enhance data recovery from fluctuating noisy modalities, we align modality distributions within a common semantic space. Extensive experiments demonstrate that our DUAL-Health outperforms state-of-the-art baselines in detection accuracy and robustness.

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Physically-Induced Atmospheric Adversarial Perturbations: Enhancing Transferability and Robustness in Remote Sensing Image Classification

    cs.CV 2026-04 unverdicted novelty 7.0 of 10

    FogFool creates fog-based adversarial perturbations using Perlin noise optimization to achieve high black-box transferability (83.74% TASR) and robustness to defenses in remote sensing classification.

  2. GAPSL: A Gradient-Aligned Parallel Split Learning over Data-Heterogeneous Edge Computing Systems

    cs.LG 2026-03 reject novelty 5.0 of 10

    GAPSL adds leader-gradient selection and direction-alignment regularization to parallel split learning, and its testbed runs show consistent accuracy and convergence-time gains over PSL, SFL, EPSL, and vanilla SL.

  3. SwarmSense-DNN: A Trustworthy and Decentralized Neural Framework for Proactive Anomaly Defense in Consumer IoT

    cs.CR 2026-06 unverdicted novelty 3.0 of 10

    SwarmSense-DNN is a proposed decentralized neural framework that integrates swarm intelligence with hierarchical federated learning and graph neural networks to achieve 95.44% anomaly detection accuracy and 67% reduce...

Reference graph

Works this paper leans on

73 extracted references · 58 canonical work pages · cited by 3 Pith papers

  1. [1]

    WHO: The top 10 causes of death. 2024. Https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death

    work page 2024

  2. [2]

    H. Wang, M. Naghavi, C. Allen, R. M. Barber, Z. A. Bhutta, A. Carter, D. C. Casey, F. J. Charlson, A. Z. Chen, M. M. Coates et al., ``Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980--2015: a systematic analysis for the global burden of disease study 2015,'' The Lancet, vol. 38...

    work page 1980

  3. [3]

    Hamza Cherif, L

    F. Hamza Cherif, L. Hamza Cherif, M. Benabdellah, and G. Nassar, `` Monitoring driver health status in real time ,'' Review of Scientific Instruments, vol. 91, no. 3, 2020

    work page 2020

  4. [4]

    Z. Qin, Y. Zhang, S. Meng, Z. Qin, and K.-K. R. Choo, ``Imaging and fusing time series for wearable sensor-based human activity recognition,'' Information Fusion, vol. 53, pp. 80--87, 2020

    work page 2020

  5. [5]

    Z. Fang, Z. Lin, S. Hu, H. Cao, Y. Deng, X. Chen, and Y. Fang, `` IC3M: In-Car Multimodal Multi-Object Monitoring for Abnormal Status of Both Driver and Passengers ,'' arXiv preprint arXiv:2410.02592, 2024

    arXiv 2024

  6. [6]

    Al-Khafajiy, T

    M. Al-Khafajiy, T. Baker, C. Chalmers, M. Asim, H. Kolivand, M. Fahim, and A. Waraich, ``Remote health monitoring of elderly through wearable sensors,'' Multimedia Tools and Applications, vol. 78, no. 17, pp. 24\,681--24\,706, 2019

    work page 2019

  7. [7]

    Y. Tang, Z. Chen, A. Li, T. Zheng, Z. Lin, J. Xu, P. Lv, Z. Sun, and Y. Gao, `` MERIT: Multimodal Wearable Vital Sign Waveform Monitoring ,'' arXiv preprint arXiv:2410.00392, 2024

    work page Pith review arXiv 2024

  8. [8]

    C. Li, C. Wong, S. Zhang, N. Usuyama, H. Liu, J. Yang, T. Naumann, H. Poon, and J. Gao, ``Llava-med: Training a large language-and-vision assistant for biomedicine in one day,'' Advances in Neural Information Processing Systems, vol. 36, 2024

    work page 2024

Show all 73 references
  1. [9]

    M. Moor, Q. Huang, S. Wu, M. Yasunaga, Y. Dalmia, J. Leskovec, C. Zakka, E. P. Reis, and P. Rajpurkar, ``Med-flamingo: a multimodal medical few-shot learner,'' in Machine Learning for Health (ML4H). 1em plus 0.5em minus 0.4em PMLR, 2023, pp. 353--367

    2023

  2. [10]

    Y. Xu, S. Biswal, S. R. Deshpande, K. O. Maher, and J. Sun, ``Raim: Recurrent attentive and intensive model of multimodal patient monitoring data,'' in Proceedings of the 24th ACM SIGKDD international conference on Knowledge Discovery & Data Mining, August 2018, pp. 2565--2573

    2018

  3. [11]

    D. Yang, S. Huang, H. Kuang, Y. Du, and L. Zhang, ``Disentangled representation learning for multimodal emotion recognition,'' in Proceedings of the 30th ACM International Conference on Multimedia, 2022, pp. 1642--1651

    2022

  4. [12]

    F. Lv, X. Chen, Y. Huang, L. Duan, and G. Lin, ``Progressive modality reinforcement for human multimodal emotion recognition from unaligned multimodal sequences,'' in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2021, pp. 2554--2562

    2021

  5. [13]

    Akbari, L

    H. Akbari, L. Yuan, R. Qian, W.-H. Chuang, S.-F. Chang, Y. Cui, and B. Gong, ``Vatt: Transformers for multimodal self-supervised learning from raw video, audio and text,'' Advances in Neural Information Processing Systems, vol. 34, pp. 24\,206--24\,221, 2021

    2021

  6. [14]

    Shi, W.-N

    B. Shi, W.-N. Hsu, K. Lakhotia, and A. Mohamed, ``Learning audio-visual speech representation by masked multimodal cluster prediction,'' arXiv preprint arXiv:2201.02184, 2022

    2022 arXiv

  7. [15]

    X. Mei, C. Meng, H. Liu, Q. Kong, T. Ko, C. Zhao, M. D. Plumbley, Y. Zou, and W. Wang, ``Wavcaps: A chatgpt-assisted weakly-labelled audio captioning dataset for audio-language multimodal research,'' IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 32, pp....

    2024

  8. [16]

    J. Gao, C. Xiao, L. M. Glass, and J. Sun, ``Compose: Cross-modal pseudo-siamese network for patient trial matching,'' in Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, August 2020, pp. 803--812

    2020

  9. [17]

    Zhang, X

    C. Zhang, X. Chu, L. Ma, Y. Zhu, Y. Wang, J. Wang, and J. Zhao, ``M3care: Learning with missing modalities in multimodal healthcare data,'' in Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, August 2022, pp. 2418--2428

    2022

  10. [18]

    S. Hu, Y. Tao, G. Xu, X. Qian, Y. Deng, X. Chen, S. T. W. Kwong, and Y. Fang, ``Cp-guard: A unified, probability-agnostic, and adaptive framework for malicious agent detection and defense in multi-agent embodied perception systems,'' arXiv preprint arXiv:2506.22890, 2025

    2025

  11. [19]

    J. Bao, H. Sun, H. Deng, Y. He, Z. Zhang, and X. Li, ``Bmad: Benchmarks for medical anomaly detection,'' in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June 2024, pp. 4042--4053

    2024

  12. [20]

    Verspoor, K

    K. Verspoor, K. B. Cohen, A. Lanfranchi, C. Warner, H. L. Johnson, C. Roeder, J. D. Choi, C. Funk, Y. Malenkiy, M. Eckert et al., ``A corpus of full-text journal articles is a robust evaluation tool for revealing differences in performance of biomedical natural language proces...

    2012

  13. [21]

    N. Chan, F. Parker, W. Bennett, T. Wu, M. Y. Jia, J. Fackler, and K. Ghobadi, ``Medtsllm: Leveraging llms for multimodal medical time series analysis,'' arXiv preprint arXiv:2408.07773, 2024

    2024 arXiv

  14. [22]

    Z. Lin, G. Qu, Q. Chen, X. Chen, Z. Chen, and K. Huang, `` Pushing Large Language Models to the 6G Edge: Vision, Challenges, and Opportunities ,'' arXiv preprint arXiv:2309.16739, 2023

    2023 arXiv

  15. [23]

    S. Hu, Z. Fang, Y. Deng, X. Chen, and Y. Fang, ``Collaborative perception for connected and autonomous driving: Challenges, possible solutions and opportunities,'' IEEE Wireless Communications, 2025

    2025

  16. [24]

    Z. Lin, Y. Zhang, Z. Chen, Z. Fang, X. Chen, P. Vepakomma, W. Ni, J. Luo, and Y. Gao, `` HSplitLoRA: A Heterogeneous Split Parameter-Efficient Fine-Tuning Framework for Large Language Models ,'' arXiv preprint arXiv:2505.02795, 2025

    2025 arXiv

  17. [25]

    Y. Kim, X. Xu, D. McDuff, C. Breazeal, and H. W. Park, ``Health-llm: Large language models for health prediction via wearable sensor data,'' arXiv preprint arXiv:2401.06866, 2024

    2024 arXiv

  18. [26]

    Z. Fang, Z. Lin, Z. Chen, X. Chen, Y. Gao, and Y. Fang, `` Automated Federated Pipeline for Parameter-Efficient Fine-Tuning of Large Language Models ,'' arXiv preprint arXiv:2404.06448, 2024

    2024 arXiv

  19. [27]

    Z. Lin, G. Qu, X. Chen, and K. Huang, `` Split Learning in 6G Edge Networks ,'' IEEE Wirel. Commun. , 2024

    2024

  20. [28]

    X. Liu, D. McDuff, G. Kovacs, I. Galatzer-Levy, J. Sunshine, J. Zhan, M.-Z. Poh, S. Liao, P. Di Achille, and S. Patel, ``Large language models are few-shot health learners,'' arXiv preprint arXiv:2305.15525, 2023

    2023 arXiv

  21. [29]

    S. Hu, Y. Ma, Y. Tao, Z. Fang, Z. Fang, Y. Deng, S. Kwong, and Y. Fang, ``Task-aware parameter-efficient fine-tuning of large pre-trained models at the edge,'' arXiv preprint arXiv:2504.03718, 2025

    2025 arXiv

  22. [30]

    A. J. Thirunavukarasu, D. S. J. Ting, K. Elangovan, L. Gutierrez, T. F. Tan, and D. S. W. Ting, ``Large language models in medicine,'' Nature Medicine, vol. 29, no. 8, pp. 1930--1940, 2023

    1930

  23. [31]

    S. Yin, C. Fu, S. Zhao, K. Li, X. Sun, T. Xu, and E. Chen, ``A survey on multimodal large language models,'' arXiv preprint arXiv:2306.13549, 2023

    2023 arXiv

  24. [32]

    J. Wang, S. Ahn, T. Dalal, X. Zhang, W. Pan, Q. Zhang, B. Chen, H. H. Dodge, F. Wang, and J. Zhou, ``Augmented risk prediction for the onset of alzheimer's disease from electronic health records with large language models,'' arXiv preprint arXiv:2405.16413, 2024

    2024 arXiv

  25. [33]

    Z. Lin, X. Hu, Y. Zhang, Z. Chen, Z. Fang, X. Chen, A. Li, P. Vepakomma, and Y. Gao, `` SplitLoRA: A Split Parameter-Efficient Fine-Tuning Framework for Large Language Models ,'' arXiv preprint arXiv:2407.00952, 2024

    2024 arXiv

  26. [34]

    S. Hu, Z. Fang, Z. Fang, Y. Deng, X. Chen, Y. Fang, and S. T. W. Kwong, ``Agentscomerge: Large language model empowered collaborative decision making for ramp merging,'' IEEE Transactions on Mobile Computing, 2025

    2025

  27. [35]

    S. Hu, Z. Fang, Z. Fang, Y. Deng, X. Chen, and Y. Fang, ``Agentscodriver: Large language model empowered collaborative driving with lifelong learning,'' arXiv preprint arXiv:2404.06345, 2024

    2024 arXiv

  28. [36]

    K. Wu, B. Jiang, Z. Jiang, Q. He, D. Luo, S. Wang, Q. Liu, and C. Wang, ``Noiseboost: Alleviating hallucination with noise perturbation for multimodal large language models,'' arXiv preprint arXiv:2405.20081, 2024

    2024 arXiv

  29. [37]

    Z. Wang, Z. Wan, and X. Wan, ``Transmodality: An end2end fusion method with transformer for multimodal sentiment analysis,'' in Proceedings of the Web Conference 2020, 2020, pp. 2514--2520

    2020

  30. [38]

    Q. Wang, L. Zhan, P. Thompson, and J. Zhou, ``Multimodal learning with incomplete modalities by knowledge distillation,'' in Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, August 2020, pp. 1828--1838

    2020

  31. [39]

    J. Zhao, R. Li, and Q. Jin, ``Missing modality imagination network for emotion recognition with uncertain missing modalities,'' in Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Lan...

    2021

  32. [40]

    K. Zhou, J. Li, Y. Xiao, J. Yang, J. Cheng, W. Liu, W. Luo, J. Liu, and S. Gao, ``Memorizing structure-texture correspondence for image anomaly detection,'' IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 6, pp. 2335--2349, 2021

    2021

  33. [41]

    Perera, R

    P. Perera, R. Nallapati, and B. Xiang, ``Ocgan: One-class novelty detection using gans with constrained latent representations,'' in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), June 2019, pp. 2898--2906

    2019

  34. [42]

    Bubeck, V

    S. Bubeck, V. Chandrasekaran, R. Eldan, J. Gehrke, E. Horvitz, E. Kamar, P. Lee, Y. T. Lee, Y. Li, S. Lundberg et al., ``Sparks of artificial general intelligence: Early experiments with gpt-4,'' arXiv preprint arXiv:2303.12712, 2023

    2023 arXiv

  35. [43]

    H. W. Chung, L. Hou, S. Longpre, B. Zoph, Y. Tay, W. Fedus, Y. Li, X. Wang, M. Dehghani, S. Brahma et al., ``Scaling instruction-finetuned language models,'' Journal of Machine Learning Research, vol. 25, no. 70, pp. 1--53, 2024

    2024

  36. [44]

    Touvron, T

    H. Touvron, T. Lavril, G. Izacard, X. Martinet, M.-A. Lachaux, T. Lacroix, B. Rozi \`e re, N. Goyal, E. Hambro, F. Azhar et al., ``Llama: Open and efficient foundation language models,'' arXiv preprint arXiv:2302.13971, 2023

    2023 arXiv

  37. [45]

    Bolton, D

    E. Bolton, D. Hall, M. Yasunaga, T. Lee, C. Manning, and P. Liang, ``Biomedlm: a domain-specific large language model for biomedical text,'' Stanford CRFM Blog, 2022

    2022

  38. [46]

    R. Luo, L. Sun, Y. Xia, T. Qin, S. Zhang, H. Poon, and T.-Y. Liu, ``Biogpt: generative pre-trained transformer for biomedical text generation and mining,'' Briefings in Bioinformatics, vol. 23, no. 6, p. bbac409, 2022

    2022

  39. [47]

    Singhal, S

    K. Singhal, S. Azizi, T. Tu, S. S. Mahdavi, J. Wei, H. W. Chung, N. Scales, A. Tanwani, H. Cole-Lewis, S. Pfohl et al., ``Large language models encode clinical knowledge,'' Nature, vol. 620, no. 7972, pp. 172--180, 2023

    2023

  40. [48]

    Nagrani, S

    A. Nagrani, S. Yang, A. Arnab, A. Jansen, C. Schmid, and C. Sun, ``Attention bottlenecks for multimodal fusion,'' Advances in Neural Information Processing Systems, vol. 34, pp. 14\,200--14\,213, 2021

    2021

  41. [49]

    Blundell, J

    C. Blundell, J. Cornebise, K. Kavukcuoglu, and D. Wierstra, `` Weight uncertainty in neural network ,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2015, pp. 1613--1622

    2015

  42. [50]

    Gal and Z

    Y. Gal and Z. Ghahramani, `` Dropout as a bayesian approximation: Representing model uncertainty in deep learning ,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2016, pp. 1050--1059

    2016

  43. [51]

    W. J. Maddox, P. Izmailov, T. Garipov, D. P. Vetrov, and A. G. Wilson, ``A simple baseline for bayesian uncertainty in deep learning,'' Advances in Neural Information Processing Systems, vol. 32, 2019

    2019

  44. [52]

    Subedar, R

    M. Subedar, R. Krishnan, P. L. Meyer, O. Tickoo, and J. Huang, ``Uncertainty-aware audiovisual activity recognition using deep bayesian variational inference,'' in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 6301--6310

    2019

  45. [53]

    Y. Ji, J. Wang, Y. Gong, L. Zhang, Y. Zhu, H. Wang, J. Zhang, T. Sakai, and Y. Yang, ``Map: Multimodal uncertainty-aware vision-language pre-training model,'' in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2023, pp. 23\,262--23\,271

    2023

  46. [54]

    Z. Gao, X. Jiang, X. Xu, F. Shen, Y. Li, and H. T. Shen, ``Embracing unimodal aleatoric uncertainty for robust multimodal fusion,'' in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2024, pp. 26\,876--26\,885

    2024

  47. [55]

    Chang, Z

    J. Chang, Z. Lan, C. Cheng, and Y. Wei, ``Data uncertainty learning in face recognition,'' in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2020, pp. 5710--5719

    2020

  48. [56]

    Kwon, J.-H

    Y. Kwon, J.-H. Won, B. J. Kim, and M. C. Paik, ``Uncertainty quantification using bayesian neural networks in classification: Application to biomedical image segmentation,'' Computational Statistics & Data Analysis, vol. 142, p. 106816, 2020

    2020

  49. [57]

    Harper and J

    R. Harper and J. Southern, ``A bayesian deep learning framework for end-to-end prediction of emotion from heartbeat,'' IEEE Transactions on Affective Computing, vol. 13, no. 2, pp. 985--991, 2020

    2020

  50. [58]

    C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, ``On calibration of modern neural networks,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2017, pp. 1321--1330

    2017

  51. [59]

    Kuleshov, N

    V. Kuleshov, N. Fenner, and S. Ermon, ``Accurate uncertainties for deep learning using calibrated regression,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2018, pp. 2796--2804

    2018

  52. [60]

    Mukhoti, V

    J. Mukhoti, V. Kulharia, A. Sanyal, S. Golodetz, P. Torr, and P. Dokania, ``Calibrating deep neural networks using focal loss,'' Advances in Neural Information Processing Systems, vol. 33, pp. 15\,288--15\,299, 2020

    2020

  53. [61]

    Sanchez, M

    E. Sanchez, M. K. Tellamekala, M. Valstar, and G. Tzimiropoulos, ``Affective processes: Stochastic modelling of temporal context for emotion and facial expression recognition,'' in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2...

    2021

  54. [62]

    S. Yao, S. Hu, Y. Zhao, A. Zhang, and T. Abdelzaher, ``Deepsense: A unified deep learning framework for time-series mobile sensing data processing,'' in Proceedings of the 26th International Conference on World Wide Web, 2017, pp. 351--360

    2017

  55. [63]

    Taamneh, P

    S. Taamneh, P. Tsiamyrtzis, M. Dcosta, P. Buddharaju, A. Khatri, M. Manser, T. Ferris, R. Wunderlich, and I. Pavlidis, ``A multimodal dataset for various forms of distracted driving,'' Scientific Data, vol. 4, no. 1, pp. 1--21, 2017

    2017

  56. [64]

    Nguyen, J

    A. Nguyen, J. Yosinski, and J. Clune, `` Deep neural networks are easily fooled: High confidence predictions for unrecognizable images ,'' in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015, pp. 427--436

    2015

  57. [65]

    J. Moon, J. Kim, Y. Shin, and S. Hwang, `` Confidence-aware learning for deep neural networks ,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2020, pp. 7034--7044

    2020

  58. [66]

    Mart \' nez-Villase \ n or, H

    L. Mart \' nez-Villase \ n or, H. Ponce, J. Brieva, E. Moya-Albor, J. N \'u \ n ez-Mart \' nez, and C. Pe \ n afort-Asturiano, `` UP-fall detection dataset: A multimodal approach ,'' Sensors, vol. 19, no. 9, p. 1988, 2019

    1988

  59. [67]

    N e mcov \'a , V

    A. N e mcov \'a , V. Svozilov \'a , K. Bucsuh \'a zy, R. Sm \' s ek, M. M \'e zl, B. Hesko, M. Bel \'a k, M. Bil \' k, P. Maxera, M. Seitl et al., ``Multimodal features for detection of driver stress and fatigue,'' IEEE Transactions on Intelligent Transportation Systems, vol. ...

    2020

  60. [68]

    W. Kim, B. Son, and I. Kim, ``Vilt: Vision-and-language transformer without convolution or region supervision,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2021, pp. 5583--5594

    2021

  61. [69]

    Z. Lin, G. Zhu, Y. Deng, X. Chen, Y. Gao, K. Huang, and Y. Fang, `` Efficient Parallel Split Learning over Resource-Constrained Wireless Edge Networks ,'' IEEE Trans. Mobile Comput. , vol. 23, no. 10, pp. 9224--9239, 2024

    2024

  62. [70]

    M. Hu, J. Zhang, X. Wang, S. Liu, and Z. Lin, `` Accelerating Federated Learning with Model Segmentation for Edge Networks ,'' IEEE Trans. Green Commun. Netw. , 2024

    2024

  63. [71]

    Zhang, H

    Y. Zhang, H. Chen, Z. Lin, Z. Chen, and J. Zhao, `` LCFed: An Efficient Clustered Federated Learning Framework for Heterogeneous Data ,'' arXiv preprint arXiv:2501.01850, 2025

    2025 arXiv

  64. [72]

    Z. Lin, G. Qu, W. Wei, X. Chen, and K. K. Leung, `` Adaptsfl: Adaptive Split Federated Learning in Resource-Constrained Edge Networks ,'' IEEE Trans. Netw. , 2024

    2024

  65. [73]

    Zhang, H

    Y. Zhang, H. Chen, Z. Lin, Z. Chen, and J. Zhao, ``Fedac: An adaptive clustered federated learning framework for heterogeneous data,'' arXiv preprint arXiv:2403.16460, 2024

    2024 arXiv

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.