Driving · 2026-10-26

Autoregressive Multimodal Anomaly Detection with Large Language Models for Microservice Systems

Yixiang Tang, Fagui Liu, Guoxiang Zhong, Xi Yao, C. L. Philip Chen

arXiv:2610.31722PDF

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

LLMMAD is an unsupervised autoregressive multimodal anomaly detection framework for microservice systems. It interleaves metrics/logs/traces into a unified sequence for LoRA-adapted, frozen LLM inference, then applies modality-wise reconstruction heads. It uses a prediction-driven autoregressive objective to predict future states (not copy inputs) to avoid the “identity shortcut,” and aggregates modality decisions with OR-logic.

Why it matters

The authors target two issues in existing multimodal anomaly detection for microservices: neglecting modality-specific anomaly cues and susceptibility to the “identity shortcut” in reconstruction-based models. LLMMAD’s unified LLM modeling with modality-wise reconstruction and its prediction-driven autoregressive objective aim to improve robustness and discriminativeness of anomaly representations for microservice monitoring.

Method

  • Time-aligns metrics, logs, and traces; then embeds each modality using 1D CNN + linear projection + layer norm, and interleaves multimodal embeddings by recombination for LLM processing.
  • Uses a LoRA-adapted LLM backbone with all LLM parameters frozen; performs inverse recombination to disentangle modality streams and applies modality-wise reconstruction heads.
  • Adopts a prediction-driven autoregressive objective that predicts future states from historical context using causal masking, decoupling inputs from reconstruction targets; aggregates decisions via OR-logic.
Abstract (from arXiv)

Microservice systems have become a dominant architecture for large-scale cloud applications, enabling rapid deployment and elastic scaling. However, ensuring the reliability and availability of microservice systems critically relies on effective anomaly detection, as failures and performance anomalies can quickly propagate across services and severely impact end-to-end system behavior. Existing multimodal anomaly detection approaches for microservice systems often struggle with two fundamental limitations: (i) the neglect of modality-specific characteristics in favor of cross-modality fusion, and (ii) the "identity-shortcut" problem that affects the feature reconstruction process. To address these issues, we propose LLMMAD, an autoregressive multimodal anomaly detection framework for microservice systems based on Large Language Models (LLMs). Initially, to reconcile cross-modality fusion with modality-specific preservation, LLMMAD embeds and recombines multimodal data into a unified sequence for LLM processing to model inter-modality relationships, while employing modality-wise reconstruction heads to retain modality-specific information. Subsequently, to eliminate the identity-shortcut, we implement a prediction-driven autoregressive objective, under which the model predicts future states rather than copies current inputs. Finally, we conduct extensive experiments on three real-world datasets. The results demonstrate that LLMMAD consistently outperforms state-of-the-art methods, achieving F1-score improvements of 3.1%, 1.4%, and 0.9%, respectively, thereby providing a more reliable solution for microservice monitoring.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-10-22 · Indexed by Pub-AI: 2026-10-26 UTC

preprint

Autoregressive Multimodal Anomaly Detection with Large Language Models for Microservice Systems

Yixiang Tang, Fagui Liu, Guoxiang Zhong, Xi Yao, C. L. Philip Chen

Preprint posted: 2026-10-22 · Original record ↗

arxiv
2610.31722
Explicit source versions (1)
  • v1 · 2026-10-22
  1. Preprint available

    Source date: 2026-10-22 · First detected: 2026-11-02 UTC · Historical baseline

    Source evidence (1)
Abstract · arxiv

Microservice systems have become a dominant architecture for large-scale cloud applications, enabling rapid deployment and elastic scaling. However, ensuring the reliability and availability of microservice systems critically relies on effective anomaly detection, as failures and performance anomalies can quickly propagate across services and severely impact end-to-end system behavior. Existing multimodal anomaly detection approaches for microservice systems often struggle with two fundamental limitations: (i) the neglect of modality-specific characteristics in favor of cross-modality fusion, and (ii) the "identity-shortcut" problem that affects the feature reconstruction process. To address these issues, we propose LLMMAD, an autoregressive multimodal anomaly detection framework for microservice systems based on Large Language Models (LLMs). Initially, to reconcile cross-modality fusion with modality-specific preservation, LLMMAD embeds and recombines multimodal data into a unified sequence for LLM processing to model inter-modality relationships, while employing modality-wise reconstruction heads to retain modality-specific information. Subsequently, to eliminate the identity-shortcut, we implement a prediction-driven autoregressive objective, under which the model predicts future states rather than copies current inputs. Finally, we conduct extensive experiments on three real-world datasets. The results demonstrate that LLMMAD consistently outperforms state-of-the-art methods, achieving F1-score improvements of 3.1%, 1.4%, and 0.9%, respectively, thereby providing a more reliable solution for microservice monitoring.