TriP-LLM: A Tri-Branch Patch-wise Large Language Model Framework for Time-Series Anomaly Detection

TL;DR

TriP-LLM introduces a novel unsupervised framework that leverages a tri-branch patch-wise approach combined with large language models to enhance time-series anomaly detection, outperforming existing methods on multiple benchmarks.

Contribution

The paper proposes a new tri-branch patch-wise LLM framework for time-series anomaly detection, integrating local and global features and demonstrating superior performance and efficiency.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Achieves lower memory consumption compared to Channel Independence approaches.

Validates the effectiveness of LLM integration through ablation studies.

Abstract

Time-series anomaly detection plays a central role across a wide range of application domains. With the increasing proliferation of the Internet of Things (IoT) and smart manufacturing, time-series data has dramatically increased in both scale and dimensionality. This growth has exposed the limitations of traditional statistical methods in handling the high heterogeneity and complexity of such data. Inspired by the recent success of large language models (LLMs) in multimodal tasks across language and vision domains, we propose a novel unsupervised anomaly detection framework: A Tri-Branch Patch-wise Large Language Model Framework for Time-Series Anomaly Detection (TriP-LLM). TriP-LLM integrates local and global temporal features through a triple-branch design comprising Patching, Selecting, and Global modules, to encode the input time-series into patch-wise representations, which are then processed by a frozen, pretrained LLM. A lightweight patch-wise decoder reconstructs the input, from which anomaly scores are derived. We evaluate TriP-LLM on several public benchmark datasets using PATE, a recently proposed threshold-free evaluation metric, and conduct all comparisons within a unified open-source framework to ensure fairness. Experimental results show that TriP-LLM consistently outperforms recent state-of-the-art (SOTA) methods across all datasets, demonstrating strong detection capabilities. Furthermore, through extensive ablation studies, we verify the substantial contribution of the LLM to the overall architecture. Compared to LLM-based approaches using Channel Independence (CI) patch processing, TriP-LLM achieves significantly lower memory consumption, making it more suitable for GPU memory-constrained environments. All code and model checkpoints of TriP-LLM are publicly available on https://github.com/YYZStart/TriP-LLM.git