Scholar
Sarah Alnegheimish
Google Scholar ID: wleaufEAAAAJ
MIT
Citations & Impact
Citations
722
H-index
8
i10-index
7
Publications
15
Co-authors
8
list available
Publications
Resume (English only)
Academic Achievements
- 1. 'Making the End-User a Priority in Benchmarking: OrionBench for Unsupervised Time Series Anomaly Detection', 2024 IEEE International Conference on Big Data.
- 2. 'Can Large Language Models be Anomaly Detectors for Time Series?', 2024 IEEE International Conference on Data Science and Advanced Analytics (DSAA 2024).
- 3. 'Using Natural Sentence Prompts for Understanding Biases in Language Models', 2022 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL).
- 4. 'Sintel: A Machine Learning Framework to Extract Insights from Signals', 2022 ACM SIGMOD International Conference on Management of Data.
Research Experience
- 1. Repurposing large language models for time series tasks, including SigLLM framework, to probe LLMs in finding anomalies in zero-shot.
- 2. Developing Orion, a machine learning library for unsupervised time series anomaly detection, offering pipelines like TadGAN and AER to identify rare patterns.
- 3. Analyzing and evaluating gender-profession bias in language models by curating a dataset of real-world natural prompts, comparing template– and natural-based prompts that elicit biased output.
Education
- Ph.D. in Electrical Engineering and Computer Science (EECS) at Massachusetts Institute of Technology (MIT), Advisor: Dr. Kalyan Veeramachaneni.
Background
- Ph.D. candidate at MIT pursuing a degree in Electrical Engineering and Computer Science (EECS), advised by Dr. Kalyan Veeramachaneni, and a member of the Data to AI Lab at LIDS. Research interest lies in the field of applied deep learning, particularly within time series data. Recently, investigating the efficacy of foundation models for unsupervised time series anomaly detection. Also interested in open-source development and building systems that make machine learning usable.
