Copertina

Decompose the model: Mechanistic interpretability in image models with Generalized Integrated Gradients (GIG)

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Kim, Yearim, Han, Sangyu, Han, Sangbum, Kwak, Nojun
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Computer Vision and Pattern Recognition Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

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

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

From Local to Global to Mechanistic: An iERF-Centered Unified Framework for Interpreting Vision Models
di: Kim, Yearim, et al.
Pubblicazione: (2026)

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

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

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

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

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

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

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

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

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

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

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

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

ConcatPlexer: Additional Dim1 Batching for Faster ViTs
di: Han, Donghoon, et al.
Pubblicazione: (2023)

MoMBS: Mixed-order minibatch sampling enhances model training from diverse-quality images
di: Li, Han, et al.
Pubblicazione: (2025)

Decomposed evaluations of geographic disparities in text-to-image models
di: Sureddy, Abhishek, et al.
Pubblicazione: (2024)

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

Cognitive resilience: Unraveling the proficiency of image-captioning models to interpret masked visual content
di: Du, Zhicheng, et al.
Pubblicazione: (2024)

ZERO: Industry-ready Vision Foundation Model with Multi-modal Prompts
di: Choi, Sangbum, et al.
Pubblicazione: (2025)

Multi-Source Collaborative Gradient Discrepancy Minimization for Federated Domain Generalization
di: Wei, Yikang, et al.
Pubblicazione: (2024)

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

Adaptively Sampling-Reusing-Mixing Decomposed Gradients to Speed Up Sharpness Aware Minimization
di: Deng, Jiaxin, et al.
Pubblicazione: (2025)

Improving Progressive Generation with Decomposable Flow Matching
di: Haji-Ali, Moayed, et al.
Pubblicazione: (2025)

An interpretable framework using foundation models for fish sex identification
di: Miao, Zheng, et al.
Pubblicazione: (2026)

Cooperative Meta-Learning with Gradient Augmentation
di: Shin, Jongyun, et al.
Pubblicazione: (2024)

DETACH : Decomposed Spatio-Temporal Alignment for Exocentric Video and Ambient Sensors with Staged Learning
di: Yoon, Junho, et al.
Pubblicazione: (2025)

TestDG: Test-time Domain Generalization for Continual Test-time Adaptation
di: Lee, Sohyun, et al.
Pubblicazione: (2025)

Integrated feature analysis for deep learning interpretation and class activation maps
di: Li, Yanli, et al.
Pubblicazione: (2024)

IG2: Integrated Gradient on Iterative Gradient Path for Feature Attribution
di: Zhuo, Yue, et al.
Pubblicazione: (2024)

APT: Improving Diffusion Models for High Resolution Image Generation with Adaptive Path Tracing
di: Han, Sangmin, et al.
Pubblicazione: (2025)

GroupCoOp: Group-robust Fine-tuning via Group Prompt Learning
di: Kim, Nayeong, et al.
Pubblicazione: (2025)

FIFO-Diffusion: Generating Infinite Videos from Text without Training
di: Kim, Jihwan, et al.
Pubblicazione: (2024)

Contrastive Integrated Gradients: A Feature Attribution-Based Method for Explaining Whole Slide Image Classification
di: Vu, Anh Mai, et al.
Pubblicazione: (2025)

DeClotH: Decomposable 3D Cloth and Human Body Reconstruction from a Single Image
di: Nam, Hyeongjin, et al.
Pubblicazione: (2025)

LighthouseGS: Indoor Structure-aware 3D Gaussian Splatting for Panorama-Style Mobile Captures
di: Han, Seungoh, et al.
Pubblicazione: (2025)

Cross-Axis Feature Fusion with Joint-Wise Motion Difference Prediction for Text-Based 3D Human Motion Editing
di: Han, Gyojin, et al.
Pubblicazione: (2026)

Think-Before-Draw: Decomposing Emotion Semantics & Fine-Grained Controllable Expressive Talking Head Generation
di: Shi, Hanlei, et al.
Pubblicazione: (2025)

A Mechanistic View on Video Generation as World Models: State and Dynamics
di: Wang, Luozhou, et al.
Pubblicazione: (2026)