Publications | eXCV Workshop at ICCV 2025 https://excv-workshop.github.io/2025/publication/ Publications Source Themes Academic (https://sourcethemes.com/academic/)en-us© 2026 eXCV Workshop OrganizersFri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/img/icon.png Publications https://excv-workshop.github.io/2025/publication/ AIM: Amending Inherent Interpretability via Self-Supervised Masking https://excv-workshop.github.io/2025/publication/aim-amending-inherent-interpretability-via-self-supervised-masking/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/aim-amending-inherent-interpretability-via-self-supervised-masking/ As large as it gets: Learning infinitely large Filters via Neural Implicit Functions in the Fourier Domain https://excv-workshop.github.io/2025/publication/as-large-as-it-gets-learning-infinitely-large-filters-via-neural-implicit-functions-in-the-fourier-domain/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/as-large-as-it-gets-learning-infinitely-large-filters-via-neural-implicit-functions-in-the-fourier-domain/ BayesSDF: Surface-Based Laplacian Uncertainty Estimation for 3D Geometry with Neural Signed Distance Fields https://excv-workshop.github.io/2025/publication/bayessdf-surface-based-laplacian-uncertainty-estimation-for-3d-geometry-with-neural-signed-distance-fields/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/bayessdf-surface-based-laplacian-uncertainty-estimation-for-3d-geometry-with-neural-signed-distance-fields/ Causal Interpretation of Sparse Autoencoder Features in Vision https://excv-workshop.github.io/2025/publication/causal-interpretation-of-sparse-autoencoder-features-in-vision/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/causal-interpretation-of-sparse-autoencoder-features-in-vision/ CoCo-Bot: Energy-based Composable Concept Bottlenecks for Interpretable Generative Models https://excv-workshop.github.io/2025/publication/coco-bot-energy-based-composable-concept-bottlenecks-for-interpretable-generative-models/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/coco-bot-energy-based-composable-concept-bottlenecks-for-interpretable-generative-models/ Controlling Neural Collapse Enhances Out-of-Distribution Detection and Transfer Learning https://excv-workshop.github.io/2025/publication/controlling-neural-collapse-enhances-out-of-distribution-detection-and-transfer-learning/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/controlling-neural-collapse-enhances-out-of-distribution-detection-and-transfer-learning/ DCBM: Data-Efficient Visual Concept Bottleneck Models https://excv-workshop.github.io/2025/publication/dcbm-data-efficient-visual-concept-bottleneck-models/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/dcbm-data-efficient-visual-concept-bottleneck-models/ Do VLMs Have Bad Eyes? Diagnosing Compositional Failures via Mechanistic Interpretability https://excv-workshop.github.io/2025/publication/do-vlms-have-bad-eyes-diagnosing-compositional-failures-via-mechanistic-interpretability/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/do-vlms-have-bad-eyes-diagnosing-compositional-failures-via-mechanistic-interpretability/ Explaining Object Detection Through Difference Map https://excv-workshop.github.io/2025/publication/explaining-object-detection-through-difference-map/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/explaining-object-detection-through-difference-map/ GFR-CAM: Gram-Schmidt Feature Reduction for Hierarchical Class Activation Maps https://excv-workshop.github.io/2025/publication/gfr-cam-gram-schmidt-feature-reduction-for-hierarchical-class-activation-maps/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/gfr-cam-gram-schmidt-feature-reduction-for-hierarchical-class-activation-maps/ Gradient-based Visual Explanation for Transformer-based CLIP https://excv-workshop.github.io/2025/publication/gradient-based-visual-explanation-for-transformer-based-clip/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/gradient-based-visual-explanation-for-transformer-based-clip/ Granular Concept Circuits: Toward a Fine-Grained Circuit Discovery for Concept Representations https://excv-workshop.github.io/2025/publication/granular-concept-circuits-toward-a-fine-grained-circuit-discovery-for-concept-representations/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/granular-concept-circuits-toward-a-fine-grained-circuit-discovery-for-concept-representations/ Hallucinatory Image Tokens: A Training-free EAZY Approach on Detecting and Mitigating Object Hallucinations in LVLMs https://excv-workshop.github.io/2025/publication/hallucinatory-image-tokens-a-training-free-eazy-approach-on-detecting-and-mitigating-object-hallucinations-in-lvlms/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/hallucinatory-image-tokens-a-training-free-eazy-approach-on-detecting-and-mitigating-object-hallucinations-in-lvlms/ Interpretable Open-Vocabulary Referring Object Detection with Reverse Contrast Attention https://excv-workshop.github.io/2025/publication/interpretable-open-vocabulary-referring-object-detection-with-reverse-contrast-attention/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/interpretable-open-vocabulary-referring-object-detection-with-reverse-contrast-attention/ Is CLIP ideal? No. Can we fix it? Yes! https://excv-workshop.github.io/2025/publication/is-clip-ideal-no.-can-we-fix-it-yes/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/is-clip-ideal-no.-can-we-fix-it-yes/ Keep the Faith: Faithful Explanations in Convolutional Neural Networks for Case-Based Reasoning https://excv-workshop.github.io/2025/publication/keep-the-faith-faithful-explanations-in-convolutional-neural-networks-for-case-based-reasoning/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/keep-the-faith-faithful-explanations-in-convolutional-neural-networks-for-case-based-reasoning/ LR0.FM: Low-Res Benchmark and Improving Robustness for Zero-Shot Classification in Foundation Models https://excv-workshop.github.io/2025/publication/lr0.fm-low-res-benchmark-and-improving-robustness-for-zero-shot-classification-in-foundation-models/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/lr0.fm-low-res-benchmark-and-improving-robustness-for-zero-shot-classification-in-foundation-models/ LucidPPN: Unambiguous Prototypical Parts Network for User-centric Interpretable Computer Vision https://excv-workshop.github.io/2025/publication/lucidppn-unambiguous-prototypical-parts-network-for-user-centric-interpretable-computer-vision/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/lucidppn-unambiguous-prototypical-parts-network-for-user-centric-interpretable-computer-vision/ Patch-wise Retrieval: An Interpretable Instance-Level Image Matching https://excv-workshop.github.io/2025/publication/patch-wise-retrieval-an-interpretable-instance-level-image-matching/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/patch-wise-retrieval-an-interpretable-instance-level-image-matching/ Rethinking Explainer Trust: A Position on the Inconsistencies of Visual Explanations in Weakly Supervised Segmentation https://excv-workshop.github.io/2025/publication/rethinking-explainer-trust-a-position-on-the-inconsistencies-of-visual-explanations-in-weakly-supervised-segmentation/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/rethinking-explainer-trust-a-position-on-the-inconsistencies-of-visual-explanations-in-weakly-supervised-segmentation/ Semantic Resonance Maps: Cross-Modal Oscillations for Explainable Vision-Language Model Interpretability https://excv-workshop.github.io/2025/publication/semantic-resonance-maps-cross-modal-oscillations-for-explainable-vision-language-model-interpretability/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/semantic-resonance-maps-cross-modal-oscillations-for-explainable-vision-language-model-interpretability/ SUB: Benchmarking CBM Generalization via Synthetic Attribute Substitutions https://excv-workshop.github.io/2025/publication/sub-benchmarking-cbm-generalization-via-synthetic-attribute-substitutions/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/sub-benchmarking-cbm-generalization-via-synthetic-attribute-substitutions/ TAB: Transformer Attention Bottlenecks enable User Intervention and Debugging in Vision-Language Models https://excv-workshop.github.io/2025/publication/tab-transformer-attention-bottlenecks-enable-user-intervention-and-debugging-in-vision-language-models/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/tab-transformer-attention-bottlenecks-enable-user-intervention-and-debugging-in-vision-language-models/ The Myth of Robust Classes: How Shielding Skews Perceived Stability https://excv-workshop.github.io/2025/publication/the-myth-of-robust-classes-how-shielding-skews-perceived-stability/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/the-myth-of-robust-classes-how-shielding-skews-perceived-stability/ Toward a Principled Theory of XAI via Spectral Analysis https://excv-workshop.github.io/2025/publication/toward-a-principled-theory-of-xai-via-spectral-analysis/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/toward-a-principled-theory-of-xai-via-spectral-analysis/ Towards Safer and Understandable Driver Intention Prediction https://excv-workshop.github.io/2025/publication/towards-safer-and-understandable-driver-intention-prediction/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/towards-safer-and-understandable-driver-intention-prediction/ Unmasking the functionality of early layers in VLMs https://excv-workshop.github.io/2025/publication/unmasking-the-functionality-of-early-layers-in-vlms/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/unmasking-the-functionality-of-early-layers-in-vlms/ Very Short and Accurate Explanations by Design https://excv-workshop.github.io/2025/publication/very-short-and-accurate-explanations-by-design/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/very-short-and-accurate-explanations-by-design/ Vision language models are blind https://excv-workshop.github.io/2025/publication/vision-language-models-are-blind/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/vision-language-models-are-blind/ Vision language models fail to translate detailed visual features into words https://excv-workshop.github.io/2025/publication/vision-language-models-fail-to-translate-detailed-visual-features-into-words/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/vision-language-models-fail-to-translate-detailed-visual-features-into-words/ VITAL: More Understandable Feature Visualization through Distribution Alignment and Relevant Information Flow https://excv-workshop.github.io/2025/publication/vital-more-understandable-feature-visualization-through-distribution-alignment-and-relevant-information-flow/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/vital-more-understandable-feature-visualization-through-distribution-alignment-and-relevant-information-flow/ What Variables Affect Out-of-Distribution Generalization in Pretrained Models? https://excv-workshop.github.io/2025/publication/what-variables-affect-out-of-distribution-generalization-in-pretrained-models/ Fri, 10 Oct 2025 22:02:03 +0200 https://excv-workshop.github.io/2025/publication/what-variables-affect-out-of-distribution-generalization-in-pretrained-models/