On Background Bias of Post-Hoc Concept Embeddings in Computer Vision DNNs

Documenti analoghi

• Explaining, Verifying, and Aligning Semantic Hierarchies in Vision-Language Model Embeddings
  di: Schwalbe, Gesina, et al.
  Pubblicazione: (2026)
• Local Concept Embeddings for Analysis of Concept Distributions in Vision DNN Feature Spaces
  di: Mikriukov, Georgii, et al.
  Pubblicazione: (2023)
• Unveiling Ontological Commitment in Multi-Modal Foundation Models
  di: Keser, Mert, et al.
  Pubblicazione: (2024)
• Benchmarking Vision Foundation Models for Input Monitoring in Autonomous Driving
  di: Keser, Mert, et al.
  Pubblicazione: (2025)
• Shape Bias and Robustness Evaluation via Cue Decomposition for Image Classification and Segmentation
  di: Heinert, Edgar, et al.
  Pubblicazione: (2025)