Beyond Heatmaps: Unsupervised Concept-Graph Reasoning for Interpretable Visual Explanation

arXiv:2607.01416 · cs.CV · Submitted 2026-07-01 · Read on arXiv

cs.CV

Submitted: 2026-07-01

Updated: 2026-10-11

Terminology

• AUC — 2× in this paper · explained in 2 episodes
• Grounding — 2× in this paper · explained in 3 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Graph Attention Network (GAT) — 1× in this paper · explained in 2 episodes

Sources

• Causally Reliable Concept Bottleneck Models
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Gaussian Error Linear Units (GELUs)
• Label-Free Concept Bottleneck Models
• RISE: Randomized Input Sampling for Explanation of Black-box Models
• SmoothGrad: removing noise by adding noise
• Graph Attention Networks
• Post-hoc Concept Bottleneck Models

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models