SOTAVerified

Disentanglement

This is an approach to solve a diverse set of tasks in a data efficient manner by disentangling (or isolating ) the underlying structure of the main problem into disjoint parts of its representations. This disentanglement can be done by focussing on the "transformation" properties of the world(main problem)

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 81–90 of 1854 papers

TitleStatusHype
TokenMotion: Decoupled Motion Control via Token Disentanglement for Human-centric Video Generation—0
Steering CLIP's vision transformer with sparse autoencoders—0
Latent Diffusion Autoencoders: Toward Efficient and Meaningful Unsupervised Representation Learning in Medical ImagingCode1
VideoSPatS: Video SPatiotemporal Splines for Disentangled Occlusion, Appearance and Motion Modeling and Editing—0
Learning Sparse Disentangled Representations for Multimodal Exclusion Retrieval—0
Efficient Model Editing with Task-Localized Sparse Fine-tuningCode0
ConMo: Controllable Motion Disentanglement and Recomposition for Zero-Shot Motion TransferCode0
EagleVision: Object-level Attribute Multimodal LLM for Remote SensingCode1
Language-Guided Trajectory Traversal in Disentangled Stable Diffusion Latent Space for Factorized Medical Image Generation—0
Unsupervised Feature Disentanglement and Augmentation Network for One-class Face Anti-spoofing—0
Show:102550
← PrevPage 9 of 186Next →

No leaderboard results yet.