mlsanityData to 7 Oct 2026

Composite Concept Extraction through Backdooring

arXiv 2406.13411 ↗First posted 19 Jun 2024cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Learning composite concepts, such as textquotedbl red cartextquotedbl , from individual examples -- like a white car representing the concept of textquotedbl cartextquotedbl and a red strawberry representing the concept of textquotedbl redtextquotedbl -- is inherently challenging. This paper introduces a novel method called Composite Concept Extractor (CoCE), which leverages techniques from traditional backdoor attacks to learn these composite concepts in a zero-shot setting, requiring only examples of individual concepts. By repurposing the trigger-based model backdooring mechanism, we create a strategic distortion in the manifold of the target object (e.g., textquotedbl cartextquotedbl ) induced by example objects with the target property (e.g., textquotedbl redtextquotedbl ) from objects textquotedbl red strawberrytextquotedbl , ensuring the distortion selectively affects the target objects with the target property. Contrastive learning is then employed to further refine this distortion, and a method is formulated for detecting objects that are influenced by the distortion. Extensive experiments with in-depth analysis across different datasets demonstrate the utility and applicability of our proposed approach.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 37 references that are in the map. First posted on arXiv 19 Jun 2024.

Builds on

29 of 37 references.

References · 29

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  3. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  4. 2018Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNscited by 585
  5. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 437
  6. 2019Hidden Trigger Backdoor Attackscited by 359
  7. 2018Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooringcited by 334
  8. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 313
  9. 2020Input-Aware Dynamic Backdoor Attackcited by 304
  10. 2018Graph R-CNN for Scene Graph Generationcited by 296
  11. 2021Adversarial Neuron Pruning Purifies Backdoored Deep Modelscited by 230
  12. 2020Backdoor Attacks Against Deep Learning Systems in the Physical Worldcited by 129
17 more
  1. 2021Open World Compositional Zero-Shot Learningcited by 100
  2. 2022Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protectioncited by 80
  3. 2022Black-box Dataset Ownership Verification via Backdoor Watermarkingcited by 57
  4. 2023Reconstructive Neuron Pruning for Backdoor Defensecited by 56
  5. 2022Siamese Contrastive Embedding Network for Compositional Zero-Shot Learningcited by 55
  6. 2015Ask Your Neurons: A Neural-based Approach to Answering Questions about Imagescited by 51
  7. 2020Towards Probabilistic Verification of Machine Unlearningcited by 46
  8. 2019Gotta Catch 'Em All: Using Honeypots to Catch Adversarial Attacks on Neural Networkscited by 40
  9. 2020Visual Concept-Metaconcept Learningcited by 39
  10. 2023Label Poisoning is All You Needcited by 27
  11. 2022Membership Inference via Backdooringcited by 24
  12. 2022FALCON: Fast Visual Concept Learning by Integrating Images, Linguistic descriptions, and Conceptual Relationscited by 20
  13. 2020Scalable Backdoor Detection in Neural Networkscited by 15
  14. 2022Towards Effective and Robust Neural Trojan Defenses via Input Filteringcited by 7
  15. 2023FreeEagle: Detecting Complex Neural Trojans in Data-Free Casescited by 7
  16. 2021Semantic Host-free Trojan Attackcited by 1
  17. 2022Defense Against Multi-target Trojan Attackscited by 1

Built on by

Not cited yet.