AI Paper Atlas

Research map / 3D vision / 3D generation

Learning to Act Robustly with View-Invariant Latent Actions

Youngjoon Jeong, Junha Chun, Taesup Kim

CVPR 2026 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 7 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
pretrain encoder with VILA (view-invariant latent actions) vs ImageNet ResNet-18 baselineDiffusion Policy · Lift unseen views (fine-tuned)success rate (%): 77.00 -> 94.70Improved
pretrain encoder with VILA vs VanillaDiffusion Policy · Coffee unseen views (fine-tuned)success rate (%): 0.35 -> 12.65Improved
use VILA vs Vanilla on real-world pick-and-placeDiffusion Policy · Real-world novel viewsaverage success rate (%): 3.33 -> 63.33Improved
use VILA vs fine-tuned VLAsreal-world pick-and-place · Real-world novel viewsaverage success rate (%): 36.67 -> 63.33Improved
add scene-level view-invariant encoders (CLASS) vs Vanilla in frozen setting on LiftDiffusion Policy · Lift unseen views (frozen)success rate (%): 47.30 -> 65.00Mixed
add image augmentation (crop, color jitter) to VanillaVanilla encoder · Square unseen views (fine-tuned)success rate (%): 8.70 -> 9.30No clear effect
add structural alignment loss on top of weighted InfoNCE (ablation)VILA · Lift unseen views (fine-tuned)unseen success rate (%): 91.70 -> 94.70Improved
add action-aware weighting to the contrastive loss (ablation)VILA · Lift unseen views (fine-tuned)unseen success rate (%): 84.30 -> 94.70Improved
add contrastive and structural losses on top of base latent-action loss (ablation)VILA · Lift unseen views (fine-tuned)unseen success rate (%): 79.50 -> 94.70Improved
replace distance-based structural loss with CKA alignment (ablation)VILA · Lift unseen views (fine-tuned)unseen success rate (%): 94.70 -> 92.00Hurt
add auxiliary action-regression head (ablation)VILA · Lift unseen views (fine-tuned)unseen success rate (%): 94.70 -> 86.70Hurt
use DTW instead of L2 for action-sequence distance (ablation)VILA · Lift unseen views (fine-tuned)unseen success rate (%): 94.70 -> 89.20Hurt

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