Towards Trustworthy Machine Learning via Detection and Removal of Harmful Outputs

  • Qizhou WANG

Student thesis: Doctoral Thesis

Abstract

Please refer to the "Abstract & TOC" file on the right side
Date of Award27 Jun 2025
Original languageEnglish
SupervisorBo HAN (Supervisor)

User-Defined Keywords

  • Trustworthy Machine Learning
  • Out-of-distribution Detection
  • Large Language Model Unlearning

Cite this

Documents

  • Full Text

    File: application/pdf, 4.99 MB

    Type: Thesis

  • Abstract & TOC

    File: application/pdf, 22.4 KB

    Type: Supplementary materials

'