# Label-Efficient Time Series Classification at Scale: A Dual-Stream OSSE-LSTM with Counterfactual Attribution

arXiv: https://arxiv.org/abs/2610.02704 · HTML: https://mlsanity.com/paper/2610.02704/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 47 references that are in the map.
- First posted on arXiv 2 Oct 2026.

## Builds on (most cited first, 22 of 22)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [SmoothGrad: removing noise by adding noise](https://mlsanity.com/paper/1706.03825/) (2017)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [The UCR Time Series Archive](https://mlsanity.com/paper/1810.07758/) (2018)
- [InceptionTime: Finding AlexNet for Time Series Classification](https://mlsanity.com/paper/1909.04939/) (2019)
- [Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline](https://mlsanity.com/paper/1611.06455/) (2016)
- [Explanations based on the Missing: Towards Contrastive Explanations with Pertinent Negatives](https://mlsanity.com/paper/1802.07623/) (2018)
- [MINIROCKET: A Very Fast (Almost) Deterministic Transform for Time Series Classification](https://mlsanity.com/paper/2012.08791/) (2020)
- [The Shattered Gradients Problem: If resnets are the answer, then what is the question?](https://mlsanity.com/paper/1702.08591/) (2017)
- [LSTM Fully Convolutional Networks for Time Series Classification](https://mlsanity.com/paper/1709.05206/) (2017)
- [Omni-Scale CNNs: a simple and effective kernel size configuration for time series classification](https://mlsanity.com/paper/2002.10061/) (2020)
- [Deep learning for time series classification](https://mlsanity.com/paper/2010.00567/) (2020)
- [Interpretable Time-series Classification on Few-shot Samples](https://mlsanity.com/paper/2006.02031/) (2020)
- [Efficient Temporal Pattern Mining in Big Time Series Using Mutual Information -- Full Version](https://mlsanity.com/paper/2010.03653/) (2020)
- [AMIC: An Adaptive Information Theoretic Method to Identify Multi-Scale Temporal Correlations in Big Time Series Data -- Accepted Version](https://mlsanity.com/paper/1906.09995/) (2019)
- [A Unified Approach for Multi-Scale Synchronous Correlation Search in Big Time Series -- Full Version](https://mlsanity.com/paper/2204.09131/) (2022)
- [Mining Seasonal Temporal Patterns in Time Series](https://mlsanity.com/paper/2206.14604/) (2022)
- [Efficient Rare Temporal Pattern Mining in Time Series](https://mlsanity.com/paper/2409.05042/) (2024)
- [Distributed Seasonal Temporal Pattern Mining](https://mlsanity.com/paper/2511.12216/) (2025)

## Abstract (arXiv)

Time series are produced continuously at enormous scale by industrial equipment, wearables, power grids, and clinical monitors, yet annotation remains manual, expensive, and expert-dependent. The binding constraint in large-scale time series analytics is therefore not data volume but label volume, and the question facing a practitioner is concrete: how many examples per class must be labeled before a classifier becomes usable? We study this question directly, in a regime where the label space is fixed and known in advance and the decision rule must be constructed from only K labeled examples per class. We propose Dual-Stream OSSE-LSTM, an episodic metric-learning framework that pairs an Omni-Scale CNN with Squeeze-and-Excitation recalibration, for multi-scale motif extraction without per-dataset kernel tuning, with a Bidirectional LSTM for global temporal context. The two streams are independently normalized and fused into a prototype-oriented embedding. Because decisions taken from a few labels must also be explainable, we introduce Counterfactual Integrated Gradients (C-IG), which attributes the prototype margin between target and opposing classes rather than an isolated classifier logit, and reuses the resulting maps as soft masks for test-time prototype refinement without updating the encoder. On 19 univariate UCR datasets, OSSE-LSTM attains the highest average accuracy and per-dataset win count at every support size, and its accuracy remains within a 0.36-point band (96.36-96.72%) across that range. Its weakest configuration still exceeding the best result any compared baseline achieves at any K (93.99%).