Skip to main navigation
Skip to search
Skip to main content
Experts@Minnesota Home
Home
Profiles
Research units
University Assets
Projects and Grants
Research output
Datasets
Press/Media
Activities
Fellowships, Honors, and Prizes
Impacts
Search by expertise, name or affiliation
A Unified Detection Framework for Inference-Stage Backdoor Defenses
Xun Xian
, Ganghua Wang
, Jayanth Srinivasa
, Ashish Kundu
,
Xuan Bi
,
Mingyi Hong
,
Jie Ding
Information and Decision Sciences
Electrical and Computer Engineering
Statistics (Twin Cities)
Research output
:
Chapter in Book/Report/Conference proceeding
›
Conference contribution
12
Scopus citations
Overview
Fingerprint
Fingerprint
Dive into the research topics of 'A Unified Detection Framework for Inference-Stage Backdoor Defenses'. Together they form a unique fingerprint.
Sort by
Weight
Alphabetically
Keyphrases
Detection Framework
100%
Backdoor Attack
100%
Unified Detection
100%
False Positive Rate
40%
Detection Power
40%
Detection Rules
40%
Backdoor
40%
Computer Vision
20%
Natural Language Processing
20%
Benchmark Dataset
20%
Application-oriented
20%
Detection Problem
20%
Effective Approach
20%
State-of-the-art Techniques
20%
Real-world Application
20%
ROC-AUC
20%
Optimal Detection
20%
Latent Representation
20%
Learning Scenario
20%
Vision-language
20%
Deep Nets
20%
Backdoor Trigger
20%
Classical Learning
20%
Specific Behavior
20%
Unified Inference
20%
Backdoor Detection
20%
Computer Science
Backdoors
100%
Backdoor Attack
80%
False Positive Rate
40%
Natural Language Processing
20%
World Application
20%
Effective Approach
20%
Computer Vision
20%