AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

VITA: Vision-to-Action Flow Matching Policy

Dechen Gao, BOQI ZHAO, Andrew Lee, Ian Chuang, Hanchu Zhou, Hang Wang, Zhe Zhao, Junshan Zhang, Iman Soltani

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Flow matching distillation · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add flow latent decoding (FLD) (ablation)VITA with action AE · ThreadNeedlesuccess rate (%): 0 -> 92Improved
replace zero-padding with learned action autoencoder latents (ablation)VITA · ThreadNeedlesuccess rate (%): 0 -> 92Improved
freeze pre-trained action autoencoder (ablation)VITA · online success and offline MSEsuccess rateHurt
add FLC alongside FLD (ablation)VITA · simulated taskssuccess rateImproved
use FLC alone instead of FLD (ablation)VITA · simulated taskssuccess rateHurt
add contrastive vision-action loss (ablation)VITA · some taskssuccess rateMixed
replace action AE with variational AE (KL term) (ablation)VITA · ThreadNeedlesuccess rateHurt
make vision encoder variational (ablation)VITA · ThreadNeedlesuccess rateHurt
use VITA instead of FM transformer with AdaLNFM · inference latency, vector latentslatency (ms/chunk): 0.3307 -> 0.2215Improved
use VITA instead of FM with cross-attentionFM · inference latency, grid latentslatency (ms/chunk): 0.5102 -> 0.2502Improved
use MLP-only FM instead of VITA (ablation)FM · PushTsuccess rate (%): 88 -> 0Hurt
use VITA instead of FMFM · SlotInsertionsuccess rate (%): 82 -> 78Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy