mlsanityData to 7 Oct 2026

Revitalizing Medical Time Series with Vision-Informed Retrieval: A Vision-Language Perspective

arXiv 2609.34652 ↗First posted 28 Sep 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Medical time series (MedTS) underpin many clinical classification tasks, yet existing methods usually represent them only as numerical sequences and underuse the morphology that is explicit in waveform inspection. To bridge this gap, we introduce Vision-Informed Retrieval (ViRe), which uses a frozen VLM-derived waveform representation as a morphology-aware Query to guide retrieval from raw numerical MedTS features. Specifically, a Vision Query is extracted using pre-trained vision-language models (VLMs) to obtain morphology-aware priors from waveform plots. A tailored attention-based cross-modal retrieval mechanism then uses the Vision Query to select morphology-relevant temporal and channel evidence from the numerical representation. ViRe demonstrates strong effectiveness against ten established baselines, yielding an overall 6.42% relative improvement over the previous state of the art across six public benchmarks. Code, training scripts, and reproducibility materials are publicly available in the GitHub Repository: https://github.com/Levi-Ackman/ViRe.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 37 of its 64 references that are in the map. First posted on arXiv 28 Sep 2026.

Builds on

37 of 64 references.

References · 37

  1. 2017Attention Is All You Needcited by 59,932
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,848
  4. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  5. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  6. 2020Reformer: The Efficient Transformercited by 1,658
  7. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
  8. 2021ViLT: Vision-and-Language Transformer Without Convolution or Region Supervisioncited by 1,224
  9. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
  10. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 906
  11. 2016Recurrent Neural Networks for Multivariate Time Series with Missing Valuescited by 548
  12. 2022Non-stationary Transformers: Exploring the Stationarity in Time Series Forecastingcited by 385
25 more
  1. 2024TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variablescited by 241
  2. 2024Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)cited by 116
  3. 2024DUET: Dual Clustering Enhanced Multivariate Time Series Forecastingcited by 104
  4. 2024TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignmentcited by 97
  5. 2021Self-Supervised Transformer for Sparse and Irregularly Sampled Multivariate Clinical Time-Seriescited by 96
  6. 2021Self-supervised representation learning from 12-lead ECG datacited by 80
  7. 2021Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysiscited by 75
  8. 2025TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecastingcited by 60
  9. 2023Contrast Everything: A Hierarchical Contrastive Framework for Medical Time-Seriescited by 59
  10. 2025Time-VLM: Exploring Multimodal Vision-Language Models for Augmented Time Series Forecastingcited by 59
  11. 2024Medformer: A Multi-Granularity Patching Transformer for Medical Time-Series Classificationcited by 53
  12. 2023Multi-resolution Time-Series Transformer for Long-term Forecastingcited by 41
  13. 2024TimeBridge: Non-Stationarity Matters for Long-term Time Series Forecastingcited by 38
  14. 2022First De-Trend then Attend: Rethinking Attention for Time-Series Forecastingcited by 19
  15. 2025FreEformer: Frequency Enhanced Transformer for Multivariate Time Series Forecastingcited by 18
  16. 2018An Automated System for Epilepsy Detection using EEG Brain Signals based on Deep Learning Approachcited by 16
  17. 2022ME-GAN: Learning Panoptic Electrocardio Representations for Multi-view ECG Synthesis Conditioned on Heart Diseasescited by 14
  18. 2024REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updatescited by 11
  19. 2024ViTime: Foundation Model for Time Series Forecasting Powered by Vision Intelligencecited by 11
  20. 2025On the Perception Bottleneck of VLMs for Chart Understandingcited by 9
  21. 2024On the Feasibility of Vision-Language Models for Time-Series Classificationcited by 8
  22. 2023Semi-Supervised Learning for Multi-Label Cardiovascular Diseases Prediction:A Multi-Dataset Studycited by 6
  23. 2024Transfer Learning in ECG Diagnosis: Is It Effective?cited by 3
  24. 2025MEETI: A Multimodal ECG Dataset from MIMIC-IV-ECG with Signals, Images, Features and Interpretationscited by 2
  25. 2025MTM: A Multi-Scale Token Mixing Transformer for Irregular Multivariate Time Series Classificationcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026MedTVL: Harnessing Vision and Language for Medical Time Series Classification
  2. 2026ViSTA: A Simple Bridge Extends Visual Alignment to Clinical Time-Series Understanding in Multimodal LLMs
  3. 2025Taming Vision-Language Models for Medical Image Analysis: A Comprehensive Review
  4. 2026Medical Context Distorts Decisions in Clinical Vision Language Models
  5. 2025Harnessing Vision Models for Time Series Analysis: A Survey
  6. 2025MMXU: A Multi-Modal and Multi-X-ray Understanding Dataset for Disease Progression
  7. 2025TemMed-Bench: Evaluating Temporal Medical Image Reasoning in Vision-Language Models
  8. 2025MOTOR: Multimodal Optimal Transport via Grounded Retrieval in Medical Visual Question Answering