Summary of Hierarchical Multimodal Llms with Semantic Space Alignment For Enhanced Time Series Classification, by Xiaoyu Tao et al.
Hierarchical Multimodal LLMs with Semantic Space Alignment for Enhanced Time Series Classification
by Xiaoyu Tao, Tingyue Pan, Mingyue Cheng, Yucong Luo
First submitted to arxiv on: 24 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes HiTime, a hierarchical multi-modal model that integrates temporal information into large language models (LLMs) for multivariate time series classification (MTSC). The model uses a hierarchical feature encoder to capture diverse aspects of time series data and a dual-view contrastive alignment module to bridge the gap between modalities. A hybrid prompting strategy is used to fine-tune the pre-trained LLM in a parameter-efficient manner. Experimental results on benchmark datasets demonstrate that HiTime achieves state-of-the-art classification performance through text generation, outperforming most competitive baseline methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary HiTime is a new way to use large language models for time series data. It’s like having two different ways of looking at the same information: one for numbers and one for words. The model uses special tricks to make sure it can understand both types of data, which helps it learn from both numbers and words together. This makes it better than other methods that only look at one type of data. |
Keywords
» Artificial intelligence » Alignment » Classification » Encoder » Multi modal » Parameter efficient » Prompting » Text generation » Time series
