Timeinteract enables real-time responses for streaming time series data
TimeInteract: Towards Real-Time Interactive Intelligence for Streaming Time Series
Machine LearningArtificial Intelligence
Summary
Time series data, like stock prices or sensor readings, keeps changing all the time, but existing models usually wait until all data is collected or pause while responding. The researchers developed TimeInteract, a system that watches data as it comes in and can decide when to respond without stopping the data flow. It uses special techniques to understand both recent changes and long-term trends while deciding when to speak up. Their tests show TimeInteract can respond faster and more accurately than previous models, making it better at handling real-time interactions with time series data.
What this means in practice
- •For financial analysts: Use TimeInteract to get continuous, real-time insights and alerts from streaming market data without waiting for data batches.
- •For industrial automation engineers: Integrate TimeInteract for immediate detection and interactive responses to changes in sensor data on factory equipment to improve maintenance scheduling.
Authors
Sheng Pan, Yongli Gu, Yiqing Guo, Warren Jin, Bo Du, Shirui Pan, Ming Jin
Abstract
Real-world time series evolve continuously, with meaningful changes potentially emerging at any moment. However, existing time-series language models (TSLMs) remain inherently static. They either receive complete sequences for offline processing or alternate between streaming input and response generation, which prevents processing of new observations during interaction. We introduce a new regime, Time-Series Interaction: a model continuously perceives incoming time-series observations and user intent, autonomously decides when to remain silent or respond, and continues processing new observations during response generation. To realize this, we develop TimeInteract with three key designs: a dual-view streaming TS encoder that captures local variations and historical dynamics, a response control mechanism that learns when to trigger a response, and a decoupled streaming inference mechanism that separates control from response generation to avoid blocking subsequent observations. We further formulate a hierarchy of interaction capabilities, progressing from Understanding to Adaptivity. Based on this hierarchy, we construct StreamTSI-34K, a large-scale streaming TS interaction dataset with 34,588 episodes and 77,505 responses across synthetic and real-world time series in single- and multi-turn settings. Across all four interaction levels, TimeInteract consistently outperforms existing LLMs, VLMs, and TSLMs, with gains of up to 23.92 points on challenging tasks. It also improves response triggering while achieving near-zero stream stall and up to $2.15\times$ inference speedup.