AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Temporal Straightening for Latent Planning

Ying Wang, Oumayma Bounou, Gaoyue Zhou, Randall Balestriero, Tim G. J. Rudner, Yann LeCun, Mengye Ren

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Probabilistic embeddings and relational inference · 0 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 curvature regularizer to a projector on frozen DINOv2 patch features (ablation)DINOv2 patch + proj 14x14x8 · PointMaze UMaze open-loop GDsuccess rate (%): 44.00 -> 94.00Improved
add curvature regularizer to ResNet trained from scratch with spatial features (ablation)ResNet 14x14x8 · Wall open-loop GDsuccess rate (%): 1.33 -> 84.67Improved
add curvature regularizer to ResNet with global features (ablation)ResNet 1x384 · PushT MPCsuccess rate (%): 70.00 -> 83.33Improved
add curvature regularizer to projector with global features (ablation)DINOv2 patch + proj 1x384 · PushTsuccess rate (%): 2.00 -> 2.00 open-loop; 11.33 -> 8.67 MPCNo clear effect
train encoder/projector with JEPA loss only (no explicit straightening) instead of frozen DINOv2 patch (ablation)DINOv2 patch + proj · Wall open-loop GDsuccess rate (%): 52.67 -> 80.00Improved
add smoothness or temporal contrastive regularizer instead of curvature (ablation)DINOv2 + spatial projector · PushT open-loop GDsuccess rate (%)No clear effect
use learnable aggregation head for curvature loss instead of patch, mean or flatten cosine (ablation)DINOv2 + spatial projector · Wall, UMaze, Medium, PushT open-loopsuccess rate (%)Improved
increase channel dimension of projected DINOv2 features from 8 to 128 (ablation)DINOv2 + spatial projector · Wall, mazes, PushT open-loopsuccess rateHurt
collapse patch features into a single global vector (ablation)DINOv2 + projector · all four environmentssuccess rateHurt
use CEM instead of GD with straightened encoder (ablation)DINOv2 patch + proj + Lcurv · PointMaze UMaze open-loopsuccess rate (%): 94.00 -> 94.00No clear effect
add global goal cost on aggregated features with weight 0.1 for long-horizon planning (ablation)Straightened models, MPC · PointMaze Medium, PushT long horizonsuccess rate (%): PushT ResNet 33.33 -> 36.00; Proj 24.00 -> 33.33Improved
add curvature regularizer to ResNet for long-horizon planning (ablation)ResNet · PointMaze Medium 50-step MPCsuccess rate (%): 48.00 -> 98.67Improved

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