BayesPrompt: Prompting Large-Scale Pre-Trained Language Models on Few-shot Inference via Debiased Domain Abstraction

1 code implementation • 25 Jan 2024 • Jiangmeng Li, Fei Song, Yifan Jin, Wenwen Qiang, Changwen Zheng, Fuchun Sun, Hui Xiong

From the perspective of distribution analyses, we disclose that the intrinsic issues behind the phenomenon are the over-multitudinous conceptual knowledge contained in PLMs and the abridged knowledge for target downstream domains, which jointly result in that PLMs mis-locate the knowledge distributions corresponding to the target domains in the universal knowledge embedding space.

Domain Adaptation

Neural-based Modeling for Performance Tuning of Spark Data Analytics

1 code implementation • 20 Jan 2021 • Khaled Zaouk, Fei Song, Chenghao Lyu, Yanlei Diao

Performance modeling of cloud data analytics is crucial for performance tuning and other critical operations in the cloud.

Exathlon: A Benchmark for Explainable Anomaly Detection over Time Series

1 code implementation • 10 Oct 2020 • Vincent Jacob, Fei Song, Arnaud Stiegler, Bijan Rad, Yanlei Diao, Nesime Tatbul

Access to high-quality data repositories and benchmarks have been instrumental in advancing the state of the art in many experimental research domains.

Anomaly Detection Time Series +1

Unsupervised Part-of-Speech Tagging in Noisy and Esoteric Domains With a Syntactic-Semantic Bayesian HMM

no code implementations • WS 2012 • William M. Darling, Michael J. Paul, Fei Song

Language Modelling Unsupervised Part-Of-Speech Tagging