Copertina

Sparse autoencoders reveal selective remapping of visual concepts during adaptation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Lim, Hyesu, Choi, Jinho, Choo, Jaegul, Schneider, Steffen
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Computer Vision and Pattern Recognition Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts
di: Choi, Jinho, et al.
Pubblicazione: (2025)

MagiCapture: High-Resolution Multi-Concept Portrait Customization
di: Hyung, Junha, et al.
Pubblicazione: (2023)

CytoSAE: Interpretable Cell Embeddings for Hematology
di: Dasdelen, Muhammed Furkan, et al.
Pubblicazione: (2025)

RL makes MLLMs see better than SFT
di: Song, Junha, et al.
Pubblicazione: (2025)

Adapting Pretrained ViTs with Convolution Injector for Visuo-Motor Control
di: Hwang, Dongyoon, et al.
Pubblicazione: (2024)

RDumb: A simple approach that questions our progress in continual test-time adaptation
di: Press, Ori, et al.
Pubblicazione: (2023)

Investigating Pre-Training Objectives for Generalization in Vision-Based Reinforcement Learning
di: Kim, Donghu, et al.
Pubblicazione: (2024)

Skip-and-Play: Depth-Driven Pose-Preserved Image Generation for Any Objects
di: Jo, Kyungmin, et al.
Pubblicazione: (2024)

Scaling Up Personalized Image Aesthetic Assessment via Task Vector Customization
di: Yun, Jooyeol, et al.
Pubblicazione: (2024)

Vector Prism: Animating Vector Graphics by Stratifying Semantic Structure
di: Yun, Jooyeol, et al.
Pubblicazione: (2025)

Inverse problem regularization with hierarchical variational autoencoders
di: Prost, Jean, et al.
Pubblicazione: (2023)

Variational Bayes image restoration with compressive autoencoders
di: Biquard, Maud, et al.
Pubblicazione: (2023)

Sparse components distinguish visual pathways & their alignment to neural networks
di: Marvi, Ammar I, et al.
Pubblicazione: (2025)

Disentangled representations via score-based variational autoencoders
di: Lyo, Benjamin S. H., et al.
Pubblicazione: (2025)

Training Spatial-Frequency Visual Prompts and Probabilistic Clusters for Accurate Black-Box Transfer Learning
di: Cho, Wonwoo, et al.
Pubblicazione: (2024)

Concept-TRAK: Understanding how diffusion models learn concepts through concept-level attribution
di: Park, Yonghyun, et al.
Pubblicazione: (2025)

Out-of-distribution evaluations of channel agnostic masked autoencoders in fluorescence microscopy
di: Hurry, Christian John, et al.
Pubblicazione: (2025)

Beta-Sigma VAE: Separating beta and decoder variance in Gaussian variational autoencoder
di: Kim, Seunghwan, et al.
Pubblicazione: (2024)

Task-conditioned adaptation of visual features in multi-task policy learning
di: Marza, Pierre, et al.
Pubblicazione: (2024)

Temporal In-Context Fine-Tuning with Temporal Reasoning for Versatile Control of Video Diffusion Models
di: Kim, Kinam, et al.
Pubblicazione: (2025)

GaussianMotion: End-to-End Learning of Animatable Gaussian Avatars with Pose Guidance from Text
di: Shim, Gyumin, et al.
Pubblicazione: (2025)

Devil is in the Detail: Towards Injecting Fine Details of Image Prompt in Image Generation via Conflict-free Guidance and Stratified Attention
di: Jo, Kyungmin, et al.
Pubblicazione: (2025)

Ground-A-Score: Scaling Up the Score Distillation for Multi-Attribute Editing
di: Chang, Hangeol, et al.
Pubblicazione: (2024)

A visualization method for data domain changes in CNN networks and the optimization method for selecting thresholds in classification tasks
di: Huang, Minzhe, et al.
Pubblicazione: (2024)

Steering CLIP's vision transformer with sparse autoencoders
di: Joseph, Sonia, et al.
Pubblicazione: (2025)

Towards Calibrated Robust Fine-Tuning of Vision-Language Models
di: Oh, Changdae, et al.
Pubblicazione: (2023)

PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor
di: Jung, Jaewon, et al.
Pubblicazione: (2024)

Good Noise Makes Good Edits: A Training-Free Diffusion-Based Video Editing with Image and Text Prompts
di: Choi, Saemee, et al.
Pubblicazione: (2025)

CoreMatching: A Co-adaptive Sparse Inference Framework with Token and Neuron Pruning for Comprehensive Acceleration of Vision-Language Models
di: Wang, Qinsi, et al.
Pubblicazione: (2025)

Enabling Region-Specific Control via Lassos in Point-Based Colorization
di: Lee, Sanghyeon, et al.
Pubblicazione: (2024)

Hyperspectral unmixing for Raman spectroscopy via physics-constrained autoencoders
di: Georgiev, Dimitar, et al.
Pubblicazione: (2024)

Sparse-to-Sparse Training of Diffusion Models
di: Oliveira, Inês Cardoso, et al.
Pubblicazione: (2025)

How can embedding models bind concepts?
di: Uselis, Arnas, et al.
Pubblicazione: (2026)

Activation Quantization of Vision Encoders Needs Prefixing Registers
di: Kim, Seunghyeon, et al.
Pubblicazione: (2025)

An attempt to generate new bridge types from latent space of variational autoencoder
di: Zhang, Hongjun
Pubblicazione: (2023)

Bones Can't Be Triangles: Accurate and Efficient Vertebrae Keypoint Estimation through Collaborative Error Revision
di: Kim, Jinhee, et al.
Pubblicazione: (2024)

DataFreeShield: Defending Adversarial Attacks without Training Data
di: Lee, Hyeyoon, et al.
Pubblicazione: (2024)

Regularized Training with Generated Datasets for Name-Only Transfer of Vision-Language Models
di: Park, Minho, et al.
Pubblicazione: (2024)

Enhancing Intrinsic Features for Debiasing via Investigating Class-Discerning Common Attributes in Bias-Contrastive Pair
di: Park, Jeonghoon, et al.
Pubblicazione: (2024)

What to Preserve and What to Transfer: Faithful, Identity-Preserving Diffusion-based Hairstyle Transfer
di: Chung, Chaeyeon, et al.
Pubblicazione: (2024)