Imagen de Portada

From Local to Global to Mechanistic: An iERF-Centered Unified Framework for Interpreting Vision Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Kim, Yearim, Han, Sangyu, Kwak, Nojun
Formato:	Preprint
Publicado:	2026
Materias:	Computer Vision and Pattern Recognition
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Causal Interpretation of Sparse Autoencoder Features in Vision
por: Han, Sangyu, et al.
Publicado: (2025)

Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)
por: Kim, Yearim, et al.
Publicado: (2024)

Respect the model: Fine-grained and Robust Explanation with Sharing Ratio Decomposition
por: Han, Sangyu, et al.
Publicado: (2024)

Bi-ICE: An Inner Interpretable Framework for Image Classification via Bi-directional Interactions between Concept and Input Embeddings
por: Hong, Jinyung, et al.
Publicado: (2024)

VDPP: Video Depth Post-Processing for Speed and Scalability
por: Yoon, Daewon, et al.
Publicado: (2026)

PedaCo-Gen: Scaffolding Pedagogical Agency in Human-AI Collaborative Video Authoring
por: Baek, Injun, et al.
Publicado: (2026)

MSG Score: Automated Video Verification for Reliable Multi-Scene Generation
por: Yoon, Daewon, et al.
Publicado: (2024)

LoGoColor: Local-Global 3D Colorization for 360° Scenes
por: Chang, Yeonjin, et al.
Publicado: (2025)

Conservative Generator, Progressive Discriminator: Coordination of Adversaries in Few-shot Incremental Image Synthesis
por: Kong, Chaerin, et al.
Publicado: (2022)

Multi-dimensional Preference Alignment by Conditioning Reward Itself
por: Jang, Jiho, et al.
Publicado: (2025)

ReFlex: Text-Guided Editing of Real Images in Rectified Flow via Mid-Step Feature Extraction and Attention Adaptation
por: Kim, Jimyeong, et al.
Publicado: (2025)

Unlocking the Potential of Unlabeled Data in Semi-Supervised Domain Generalization
por: Lee, Dongkwan, et al.
Publicado: (2025)

CSF: Black-box Fingerprinting via Compositional Semantics for Text-to-Image Models
por: Lee, Junhoo, et al.
Publicado: (2026)

Mitigating the Bias in the Model for Continual Test-Time Adaptation
por: Chung, Inseop, et al.
Publicado: (2024)

Scale Alone Does not Improve Mechanistic Interpretability in Vision Models
por: Zimmermann, Roland S., et al.
Publicado: (2023)

What's Making That Sound Right Now? Video-centric Audio-Visual Localization
por: Choi, Hahyeon, et al.
Publicado: (2025)

DivCon-NeRF: Diverse and Consistent Ray Augmentation for Few-Shot NeRF
por: Lee, Ingyun, et al.
Publicado: (2025)

Harmonizing Visual and Textual Embeddings for Zero-Shot Text-to-Image Customization
por: Song, Yeji, et al.
Publicado: (2024)

Real-Time Intuitive AI Drawing System for Collaboration: Enhancing Human Creativity through Formal and Contextual Intent Integration
por: Song, Jookyung, et al.
Publicado: (2025)

The Role of Teacher Calibration in Knowledge Distillation
por: Kim, Suyoung, et al.
Publicado: (2025)

Coreset Selection for Object Detection
por: Lee, Hojun, et al.
Publicado: (2024)

MERLIN: Multimodal Embedding Refinement via LLM-based Iterative Navigation for Text-Video Retrieval-Rerank Pipeline
por: Han, Donghoon, et al.
Publicado: (2024)

ReSpinQuant: Efficient Layer-Wise LLM Quantization via Subspace Residual Rotation Approximation
por: Kim, Suyoung, et al.
Publicado: (2026)

4DGS360: 360° Gaussian Reconstruction of Dynamic Objects from a Single Video
por: Jang, Jae Won, et al.
Publicado: (2026)

ROODI: Reconstructing Occluded Objects with Denoising Inpainters
por: Chang, Yeonjin, et al.
Publicado: (2025)

Point-to-Point: Sparse Motion Guidance for Controllable Video Editing
por: Song, Yeji, et al.
Publicado: (2025)

Unleash the Potential of CLIP for Video Highlight Detection
por: Han, Donghoon, et al.
Publicado: (2024)

Unifying Global and Local Scene Entities Modelling for Precise Action Spotting
por: Tran, Kim Hoang, et al.
Publicado: (2024)

Causal Tracing of Object Representations in Large Vision Language Models: Mechanistic Interpretability and Hallucination Mitigation
por: Li, Qiming, et al.
Publicado: (2025)

ARC-NeRF: Area Ray Casting for Broader Unseen View Coverage in Few-shot Object Rendering
por: Seo, Seunghyeon, et al.
Publicado: (2024)

LLM-Powered Flood Depth Estimation from Social Media Imagery: A Vision-Language Model Framework with Mechanistic Interpretability for Transportation Resilience
por: Fuad, Nafis, et al.
Publicado: (2026)

S3D: Sketch-Driven 3D Model Generation
por: Song, Hail, et al.
Publicado: (2025)

Unifying Global-Local Representations in Salient Object Detection with Transformer
por: Ren, Sucheng, et al.
Publicado: (2021)

When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language Models
por: Choi, Jiho, et al.
Publicado: (2026)

Style Composition within Distinct LoRA modules for Traditional Art
por: Lee, Jaehyun, et al.
Publicado: (2025)

A Revisit to the Decoder for Camouflaged Object Detection
por: Ko, Seung Woo, et al.
Publicado: (2025)

Multi-Modal Interpretability for Enhanced Localization in Vision-Language Models
por: Imran, Muhammad, et al.
Publicado: (2025)

Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models
por: Che, Liwei, et al.
Publicado: (2026)

i-WiViG: Interpretable Window Vision GNN
por: Obadic, Ivica, et al.
Publicado: (2025)

GIFT: A Framework Towards Global Interpretable Faithful Textual Explanations of Vision Classifiers
por: Zablocki, Éloi, et al.
Publicado: (2024)