TY - GEN
T1 - Large Language Models for Anomaly Detection in Computational Workflows
T2 - 2024 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2024
AU - Jin, Hongwei
AU - Papadimitriou, George
AU - Raghavan, Krishnan
AU - Zuk, Pawel
AU - Balaprakash, Prasanna
AU - Wang, Cong
AU - Mandal, Anirban
AU - Deelman, Ewa
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Anomaly detection in computational workflows is critical for ensuring system reliability and security. However, traditional rule-based methods struggle to detect novel anomalies. This paper leverages large language models (LLMs) for workflow anomaly detection by exploiting their ability to learn complex data patterns. Two approaches are investigated: (1) supervised fine-tuning (SFT), where pretrained LLMs are fine-tuned on labeled data for sentence classification to identify anomalies, and (2) in-context learning (ICL), where prompts containing task descriptions and examples guide LLMs in few-shot anomaly detection without fine-tuning. The paper evaluates the performance, efficiency, and generalization of SFT models and explores zeroshot and few-shot ICL prompts and interpretability enhancement via chain-of-thought prompting. Experiments across multiple workflow datasets demonstrate the promising potential of LLMs for effective anomaly detection in complex executions.
AB - Anomaly detection in computational workflows is critical for ensuring system reliability and security. However, traditional rule-based methods struggle to detect novel anomalies. This paper leverages large language models (LLMs) for workflow anomaly detection by exploiting their ability to learn complex data patterns. Two approaches are investigated: (1) supervised fine-tuning (SFT), where pretrained LLMs are fine-tuned on labeled data for sentence classification to identify anomalies, and (2) in-context learning (ICL), where prompts containing task descriptions and examples guide LLMs in few-shot anomaly detection without fine-tuning. The paper evaluates the performance, efficiency, and generalization of SFT models and explores zeroshot and few-shot ICL prompts and interpretability enhancement via chain-of-thought prompting. Experiments across multiple workflow datasets demonstrate the promising potential of LLMs for effective anomaly detection in complex executions.
KW - anomaly detection
KW - computational workflows
KW - in-context learning
KW - large language models
KW - supervised fine-tuning
UR - http://www.scopus.com/inward/record.url?scp=85214984612&partnerID=8YFLogxK
U2 - 10.1109/SC41406.2024.00098
DO - 10.1109/SC41406.2024.00098
M3 - Conference contribution
AN - SCOPUS:85214984612
T3 - International Conference for High Performance Computing, Networking, Storage and Analysis, SC
BT - Proceedings of SC 2024
PB - IEEE Computer Society
Y2 - 17 November 2024 through 22 November 2024
ER -