# Aionoscope: Debugging Latent-State Accessibility in Time-Series Representations

arXiv: https://arxiv.org/abs/2607.00956 · HTML: https://mlsanity.com/paper/2607.00956/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 20 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Builds on (most cited first, 17 of 17)

- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [Time-Series Representation Learning via Temporal and Contextual Contrasting](https://mlsanity.com/paper/2106.14112/) (2021)
- [Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding](https://mlsanity.com/paper/2106.00750/) (2021)
- [GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation](https://mlsanity.com/paper/2410.10393/) (2024)
- [Context is Key: A Benchmark for Forecasting with Essential Textual Information](https://mlsanity.com/paper/2410.18959/) (2024)
- [ECG-QA: A Comprehensive Question Answering Dataset Combined With Electrocardiogram](https://mlsanity.com/paper/2306.15681/) (2023)
- [Rethinking Evaluation in the Era of Time Series Foundation Models: (Un)known Information Leakage Challenges](https://mlsanity.com/paper/2510.13654/) (2025)
- [CauKer: Classification Time Series Foundation Models Can Be Pretrained on Synthetic Data](https://mlsanity.com/paper/2508.02879/) (2025)
- [Pre-trained Forecasting Models: Strong Zero-Shot Feature Extractors for Time Series Classification](https://mlsanity.com/paper/2510.26777/) (2025)
- [Learning Disentangled Representations for Time Series](https://mlsanity.com/paper/2105.08179/) (2021)
- [Are Synthetic Time-series Data Really not as Good as Real Data?](https://mlsanity.com/paper/2402.00607/) (2024)
- [How Foundational are Foundation Models for Time Series Forecasting?](https://mlsanity.com/paper/2510.00742/) (2025)
- [Frequency Matters: When Time Series Foundation Models Fail Under Spectral Shift](https://mlsanity.com/paper/2511.05619/) (2025)

## Built on by (most cited first, 2 of 2)

- [Dynamical Parameters: An Interpretability Framework for Time-Series Foundation Models](https://mlsanity.com/paper/2609.34316/) (2026)
- [When, Not How Much: Evaluating Time-Series Foundation Models on Sparse Events](https://mlsanity.com/paper/2609.39386/) (2026)

## Abstract (arXiv)

Time-series models are often evaluated by what they can forecast or classify, but those scores do not show whether their representations preserve the process state a user may want to inspect: event timing, phase, amplitude, frequency, or regime variables. We introduce Aionoscope, a generator-based diagnostic tool for debugging latent-state accessibility in frozen time-series representations. Aionoscope separates process generation from observation rendering, producing seeded synthetic streams with exact categorical and dense labels across mixture complexity and nuisance variation. We instantiate Aionoscope as Primitive Process Mixtures and evaluate 37 model-plus-adapter systems with a common pooled linear-probe protocol. The main result is a mismatch between coarse and fine-grained accessibility. Most systems make component presence easy to recover, but expose dense process state much less reliably: the highest observed dense-probe row reaches 0.689 mean masked R2, while a dense-feature oracle reaches 0.999. This is the failure mode Aionoscope is designed to surface: a representation can look informative at the level of "what kind of signal is present" while hiding the timing, phase, amplitude, frequency, or regime variables needed for debugging.