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

Showing 110 of 1854 papers

TitleStatusHype
Ditto: Motion-Space Diffusion for Controllable Realtime Talking Head SynthesisCode11
Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from BackboneCode6
Versatile Diffusion: Text, Images and Variations All in One Diffusion ModelCode6
UniK3D: Universal Camera Monocular 3D EstimationCode4
ControlVAE: Tuning, Analytical Properties, and Performance AnalysisCode4
Sigmoid Loss for Language Image Pre-TrainingCode3
DEADiff: An Efficient Stylization Diffusion Model with Disentangled RepresentationsCode3
UCF: Uncovering Common Features for Generalizable Deepfake DetectionCode3
CausalVAE: Structured Causal Disentanglement in Variational AutoencoderCode2
BlendFace: Re-designing Identity Encoders for Face-SwappingCode2
Show:102550
← PrevPage 1 of 186Next →

No leaderboard results yet.