# Rondo: Unsupervised Discovery of Recurring Temporal Structure

arXiv: https://arxiv.org/abs/2609.32500 · HTML: https://mlsanity.com/paper/2609.32500/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 56 references that are in the map.
- First posted on arXiv 26 Sep 2026.

## Builds on (most cited first, 21 of 21)

- [Temporal Convolutional Networks for Action Segmentation and Detection](https://mlsanity.com/paper/1611.05267/) (2016)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding](https://mlsanity.com/paper/2106.00750/) (2021)
- [Generalized Category Discovery](https://mlsanity.com/paper/2201.02609/) (2022)
- [Parametric Classification for Generalized Category Discovery: A Baseline Study](https://mlsanity.com/paper/2211.11727/) (2022)
- [Segmental Spatiotemporal CNNs for Fine-grained Action Segmentation](https://mlsanity.com/paper/1602.02995/) (2016)
- [TimesURL: Self-supervised Contrastive Learning for Universal Time Series Representation Learning](https://mlsanity.com/paper/2312.15709/) (2023)
- [Unsupervised Action Segmentation by Joint Representation Learning and Online Clustering](https://mlsanity.com/paper/2105.13353/) (2021)
- [Toeplitz Inverse Covariance-Based Clustering of Multivariate Time Series Data](https://mlsanity.com/paper/1706.03161/) (2017)
- [TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders](https://mlsanity.com/paper/2303.00320/) (2023)
- [MetaGCD: Learning to Continually Learn in Generalized Category Discovery](https://mlsanity.com/paper/2308.11063/) (2023)
- [Temporally Consistent Unbalanced Optimal Transport for Unsupervised Action Segmentation](https://mlsanity.com/paper/2404.01518/) (2024)
- [Learning Options via Compression](https://mlsanity.com/paper/2212.04590/) (2022)
- [Hierarchical Vector Quantization for Unsupervised Action Segmentation](https://mlsanity.com/paper/2412.17640/) (2024)
- [Happy: A Debiased Learning Framework for Continual Generalized Category Discovery](https://mlsanity.com/paper/2410.06535/) (2024)
- [MantisV2: Closing the Zero-Shot Gap in Time Series Classification with Synthetic Data and Test-Time Strategies](https://mlsanity.com/paper/2602.17868/) (2026)
- [Activity Grammars for Temporal Action Segmentation](https://mlsanity.com/paper/2312.04266/) (2023)
- [E2USD: Efficient-yet-effective Unsupervised State Detection for Multivariate Time Series](https://mlsanity.com/paper/2402.14041/) (2024)
- [Online Continuous Generalized Category Discovery](https://mlsanity.com/paper/2408.13492/) (2024)
- [CLaP -- State Detection from Time Series](https://mlsanity.com/paper/2504.01783/) (2025)
- [Looking into the Unknown: Exploring Action Discovery for Segmentation of Known and Unknown Actions](https://mlsanity.com/paper/2508.05529/) (2025)

## Abstract (arXiv)

Many real-world time series data exhibit structural properties at multiple scales, from short, recurring units to complex sequences composed of these units. Unsupervised discovery of both these components and structure enables the design of intelligent systems that help interpret temporal data, thereby limiting the amount of costly human annotations required. Existing modeling approaches typically overlook the hierarchical structure inherent to many time series, treating recurring patterns at different temporal scales as independent structures. Moreover, most assume access to the complete data sequence and treat discovery as a static process, limiting their ability to evolve as new observations arrive. We introduce Rondo, an unsupervised approach for modeling recurring hierarchical structure in continuous temporal streams. By explicitly constructing vocabularies of reusable units and their recurring compositions, Rondo captures structure shared across complex temporal patterns while refining and expanding its discoveries as the stream evolves. Evaluations on temporal sequences spanning diverse domains and data modalities show that Rondo outperforms existing unsupervised recurrence-discovery baselines, with particularly pronounced advantages in limited-data and continual-stream settings. These capabilities provide a stronger foundation for recurring-pattern discovery, scalable behavior understanding, and adaptive intelligent systems operating on long, unlabeled temporal streams.