Imagen de Portada

What is Missing? Explaining Neurons Activated by Absent Concepts

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Hesse, Robin, Schaub-Meyer, Simone, Hesse, Janina, Schiele, Bernt, Roth, Stefan
Formato:	Preprint
Publicado:	2026
Materias:	Computer Vision and Pattern Recognition Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Beyond Accuracy: What Matters in Designing Well-Behaved Image Classification Models?
por: Hesse, Robin, et al.
Publicado: (2025)

Benchmarking the Attribution Quality of Vision Models
por: Hesse, Robin, et al.
Publicado: (2024)

Interpretability Without Tradeoffs: Disentangling Polysemanticity At Equal Predictive Performance
por: Bağcı, Doğukan, et al.
Publicado: (2026)

Disentangling Polysemantic Channels in Convolutional Neural Networks
por: Hesse, Robin, et al.
Publicado: (2025)

Activation Subspaces for Out-of-Distribution Detection
por: Zöngür, Barış, et al.
Publicado: (2025)

Evaluating Object-Centric Models beyond Object Discovery
por: Singh, Krishnakant, et al.
Publicado: (2026)

GLASS: Guided Latent Slot Diffusion for Object-Centric Learning
por: Singh, Krishnakant, et al.
Publicado: (2024)

FaCT: Faithful Concept Traces for Explaining Neural Network Decisions
por: Parchami-Araghi, Amin, et al.
Publicado: (2025)

Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery
por: Rao, Sukrut, et al.
Publicado: (2024)

Good Teachers Explain: Explanation-Enhanced Knowledge Distillation
por: Parchami-Araghi, Amin, et al.
Publicado: (2024)

Optimising for Interpretability: Convolutional Dynamic Alignment Networks
por: Böhle, Moritz, et al.
Publicado: (2021)

Towards Better Understanding Attribution Methods
por: Rao, Sukrut, et al.
Publicado: (2022)

DWDN: Deep Wiener Deconvolution Network for Non-Blind Image Deblurring
por: Dong, Jiangxin, et al.
Publicado: (2021)

CFM: Language-aligned Concept Foundation Model for Vision
por: Wittenmayer, Kai, et al.
Publicado: (2026)

Adaptive Hierarchical Certification for Segmentation using Randomized Smoothing
por: Anani, Alaa, et al.
Publicado: (2024)

Better Understanding Differences in Attribution Methods via Systematic Evaluations
por: Rao, Sukrut, et al.
Publicado: (2023)

Adversarial Training against Location-Optimized Adversarial Patches
por: Rao, Sukrut, et al.
Publicado: (2020)

Spatial Reasoners for Continuous Variables in Any Domain
por: Pogodzinski, Bart, et al.
Publicado: (2025)

Spatial Reasoning with Denoising Models
por: Wewer, Christopher, et al.
Publicado: (2025)

Test-Time Visual In-Context Tuning
por: Xie, Jiahao, et al.
Publicado: (2025)

Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model
por: Endres, Jannik, et al.
Publicado: (2025)

Pixel-level Certified Explanations via Randomized Smoothing
por: Anani, Alaa, et al.
Publicado: (2025)

B-cosification: Transforming Deep Neural Networks to be Inherently Interpretable
por: Arya, Shreyash, et al.
Publicado: (2024)

From Neural Activations to Concepts: A Survey on Explaining Concepts in Neural Networks
por: Lee, Jae Hee, et al.
Publicado: (2023)

Studying How to Efficiently and Effectively Guide Models with Explanations
por: Rao, Sukrut, et al.
Publicado: (2023)

Boosting Unsupervised Semantic Segmentation with Principal Mask Proposals
por: Hahn, Oliver, et al.
Publicado: (2024)

Removing Cost Volumes from Optical Flow Estimators
por: Kiefhaber, Simon, et al.
Publicado: (2025)

FastCAV: Efficient Computation of Concept Activation Vectors for Explaining Deep Neural Networks
por: Schmalwasser, Laines, et al.
Publicado: (2025)

ELODI: Ensemble Logit Difference Inhibition for Positive-Congruent Training
por: Zhao, Yue, et al.
Publicado: (2022)

AnyUp: Universal Feature Upsampling
por: Wimmer, Thomas, et al.
Publicado: (2025)

Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking
por: Deng, Kaiyuan, et al.
Publicado: (2026)

Efficient Masked Attention Transformer for Few-Shot Classification and Segmentation
por: Carrión-Ojeda, Dustin, et al.
Publicado: (2025)

Explaining the Impact of Training on Vision Models via Activation Clustering
por: Boubekki, Ahcène, et al.
Publicado: (2024)

MEt3R: Measuring Multi-View Consistency in Generated Images
por: Asim, Mohammad, et al.
Publicado: (2025)

MUFASA: A Multi-Layer Framework for Slot Attention
por: Bock, Sebastian, et al.
Publicado: (2026)

Is Synthetic Data all We Need? Benchmarking the Robustness of Models Trained with Synthetic Images
por: Singh, Krishnakant, et al.
Publicado: (2024)

PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding
por: Kuzucu, Selim, et al.
Publicado: (2026)

Fractional Concepts in Neural Networks: Enhancing Activation Functions
por: Alijani, Zahra, et al.
Publicado: (2023)

Temporal Concept Dynamics in Diffusion Models via Prompt-Conditioned Interventions
por: Gorgun, Ada, et al.
Publicado: (2025)

What Linear Probes Miss: Multi-View Probing for Weight-Space Learning
por: Heo, Eunwoo, et al.
Publicado: (2026)