SenTSR-Bench: Thinking with Injected Knowledge for Time-Series Reasoning

TL;DR

This paper introduces SenTSR-Bench, a hybrid framework combining large language models and time-series models with knowledge injection and reinforcement learning, significantly improving diagnostic reasoning in industrial time-series data.

Contribution

The paper presents a novel hybrid knowledge-injection framework and a new benchmark for time-series reasoning, enhancing model reasoning capabilities with domain-specific knowledge.

Findings

Outperforms TSLMs by 9.1%-26.1% on diagnostic tasks.

Outperforms GRLMs by 7.9%-22.4% on diagnostic tasks.

Introduces SenTSR-Bench, a real-world industrial time-series reasoning benchmark.

Abstract

Time-series diagnostic reasoning is essential for many applications, yet existing solutions face a persistent gap: general reasoning large language models (GRLMs) possess strong reasoning skills but lack the domain-specific knowledge to understand complex time-series patterns. Conversely, fine-tuned time-series LLMs (TSLMs) understand these patterns but lack the capacity to generalize reasoning for more complicated questions. To bridge this gap, we propose a hybrid knowledge-injection framework that injects TSLM-generated insights directly into GRLM's reasoning trace, thereby achieving strong time-series reasoning with in-domain knowledge. As collecting data for knowledge injection fine-tuning is costly, we further leverage a reinforcement learning-based approach with verifiable rewards (RLVR) to elicit knowledge-rich traces without human supervision, then transfer such an in-domain thinking trace into GRLM for efficient knowledge injection. We further release SenTSR-Bench, a multivariate time-series-based diagnostic reasoning benchmark collected from real-world industrial operations. Across SenTSR-Bench and other public datasets, our method consistently surpasses TSLMs by 9.1%-26.1% and GRLMs by 7.9%-22.4%, delivering robust, context-aware time-series diagnostic insights.