Institution profile

Lenovo

Industry researchasia · cn
Official website
Research library77linked papers
Opportunities0open roles
Selected work

Representative Papers

Dual-Forecaster: A Multimodal Time Series Model Integrating Descriptive and Predictive Texts

May 02, 2025

Existing unimodal time series models rely solely on numerical data, suffering from semantic sparsity; while multimodal approaches incorporate textual information, they typically leverage only unidirectional text—either historical or future—and lack fine-grained modeling of text–time semantics, temporal dynamics, and causal relationships. To address these limitations, we propose a bidirectional text-driven forecasting paradigm that jointly integrates descriptive historical text and predictive future text for the first time. We design a three-stage cross-modal alignment module—encompassing semantic, temporal, and causal alignment—leveraging a large language model for text encoding and a dedicated time series feature extractor. Extensive experiments across 15 multivariate time series benchmarks demonstrate that our method consistently matches or surpasses state-of-the-art approaches, validating the substantial performance gains enabled by bidirectional textual integration.

1 citationsRead paper
Recent publications

Latest Papers

KC-Agent: A Dual-Process Cognitive Architecture for Efficient ML Model Improvement

Aug 03, 2026

This work addresses the degradation of model performance in production environments caused by data drift by proposing KC-Agent, a continual learning agent grounded in a dual-process cognitive architecture. Integrating rapid pattern recognition (System 1) with deliberate incremental updates (System 2), KC-Agent leverages structured memory to reuse historically successful strategies and incorporates atomic changes with rollback mechanisms to ensure reliability. Empirical evaluation demonstrates that KC-Agent achieves an average accuracy of 76.8% across five datasets with an inference time of only 13.2 seconds, significantly outperforming baseline methods such as CodeAct and ToT. Furthermore, its knowledge integration mechanism yields a 91% speedup, and large language model consensus scoring rates its output quality at 8.33 out of 10.

0 citationsRead paper