Moderate-fittingasaNaturalBackdoorDefender forPre-trainedLanguageModels

Feb-7-2026, 08:05:16 GMT–Neural Information Processing Systems

Backdoor attack, which poisons asmall number of training samples by inserting backdoor triggers, is a typical threat to security.

machine learning, model capacity, natural language, (17 more...)

Neural Information Processing Systems

Feb-7-2026, 08:05:16 GMT

Conferences PDF

Country:
- Asia
  - China
    - Beijing > Beijing (0.05)
    - Shanghai > Shanghai (0.04)
    - Zhejiang Province > Hangzhou (0.04)
  - Nepal (0.04)
- Europe
  - Ireland > Leinster
    - County Dublin > Dublin (0.04)
  - Italy > Sicily
    - Palermo (0.04)
- North America
  - Canada > Quebec
    - Montreal (0.04)
  - Dominican Republic (0.04)
  - United States
    - California > San Diego County
      - San Diego (0.04)
    - Louisiana (0.04)
    - Washington > King County
      - Seattle (0.04)
- Oceania > Australia
  - New South Wales > Sydney (0.04)

Genre:
- Research Report (0.68)

Industry:
- Information Technology > Security & Privacy (0.35)

Technology:
- Information Technology > Artificial Intelligence
  - Machine Learning (1.00)
  - Natural Language (1.00)

Duplicate Docs Excel Report

Title
Moderate-fitting as a Natural Backdoor Defender for Pre-trained Language Models

Similar Docs Excel Report more

Title	Similarity	Source
None found