TelecomTS: A Multi-Modal Observability Dataset for Time Series and Language Analysis

2025-10-07

AI Summary

Existing observability datasets are typically anonymized and normalized, discarding magnitude information and domain semantics—hindering tasks such as anomaly detection, root-cause analysis, and multimodal reasoning. To address this, we introduce the first large-scale, de-anonymized, multimodal observability dataset sourced directly from a production 5G telecommunications network. It fully preserves the absolute scale, physical units, and semantic meaning of covariates. The dataset enables joint modeling across time-series forecasting, anomaly localization, root-cause inference, and cross-modal question answering. Empirical evaluation reveals that state-of-the-art models suffer substantial performance degradation on high-noise, highly nonstationary industrial time series—demonstrating the critical need for explicit magnitude-aware modeling. This work establishes the first foundational benchmark for time-series–language joint modeling grounded in real-world industrial dimensions and semantics.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Computer Vision: Multi-modal Vision

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Modern enterprises generate vast streams of time series metrics when monitoring complex systems, known as observability data. Unlike conventional time series from domains such as weather, observability data are zero-inflated, highly stochastic, and exhibit minimal temporal structure. Despite their importance, observability datasets are underrepresented in public benchmarks due to proprietary restrictions. Existing datasets are often anonymized and normalized, removing scale information and limiting their use for tasks beyond forecasting, such as anomaly detection, root-cause analysis, and multi-modal reasoning. To address this gap, we introduce TelecomTS, a large-scale observability dataset derived from a 5G telecommunications network. TelecomTS features heterogeneous, de-anonymized covariates with explicit scale information and supports a suite of downstream tasks, including anomaly detection, root-cause analysis, and a question-answering benchmark requiring multi-modal reasoning. Benchmarking state-of-the-art time series, language, and reasoning models reveals that existing approaches struggle with the abrupt, noisy, and high-variance dynamics of observability data. Our experiments also underscore the importance of preserving covariates' absolute scale, emphasizing the need for foundation time series models that natively leverage scale information for practical observability applications.

Problem

• Addressing the scarcity of public observability datasets with real-world scale information
• Enabling multi-modal tasks like anomaly detection and root-cause analysis
• Overcoming limitations of existing models on noisy, high-variance observability data

Innovation

• Large-scale observability dataset from 5G network
• De-anonymized covariates with explicit scale information
• Multi-modal reasoning support for downstream tasks