AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Learning to See Through a Baby's Eyes: Early Visual Diets Enable Robust Visual Intelligence in Humans and Machines

Yusen Cai, Qing Lin, Bhargava Satya Nunna, Mengmi Zhang

CVPR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Human-like visual cognition in VLMs · 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
use the grayscale-to-color staged order instead of shuffled order (ablation)CDiet (SimCLR-ResNet) · CO3D-C (CO3D-10, SimCLR-ResNet pretrained on CO3D)mCE: 87.3 -> 84.8Improved
use the grayscale-to-color staged order instead of reversed order (ablation)CDiet (SimCLR-ResNet) · CO3D-C (CO3D-10, SimCLR-ResNet pretrained on CO3D)mCE: 94.1 -> 84.8Improved
add early color-degraded stages instead of full-color only (ablation)CDiet (SimCLR-ResNet) · CO3D-C (CO3D-10, SimCLR-ResNet pretrained on CO3D)mCE: 90.8 -> 84.8Improved
integrate color, acuity and temporality diets (ablation)CATDiet (SimCLR-ResNet) · CO3D-C (CO3D-10, SimCLR-ResNet pretrained on CO3D)mCE: 84.8 -> 72.0Improved
add temporal continuity objective over non-smooth baseline (ablation)TDiet (SimCLR-ResNet) · CO3D-C (CO3D-10, SimCLR-ResNet pretrained on CO3D)mCEImproved
replace temporally adjacent real frames with warped-only positives (Throughout) (ablation)CombDiet (SimCLR-ResNet) · CO3D-10accuracy: 83.0 -> 70.6Hurt
replace some real adjacent frames with warped views (Adjacent) (ablation)CombDiet (SimCLR-ResNet) · CO3D-10accuracy: 83.0 -> 82.4Hurt
halve or double temporal stride (ablation)CombDiet (SimCLR-ResNet) · CO3D-10accuracy: 83.0 -> 83.3No clear effect
lengthen the CATDiet warm-up from 30% to 70% (ablation)CombDiet (SimCLR-ViT) · CO3D-10accuracy: 74.4 -> 73.1Hurt
use CombDiet instead of standard SSLSimCLR-ViT · SAYCam testaccuracy: 48.8 -> 54.5Improved
use CombDiet instead of standard SSLDINO-ViT · 3D-PC depth orderdAcc: 74.5 -> 78.9Improved
use ordered CombDiet instead of shuffled order (ablation)DINO-ResNet · SAYCam testaccuracy: 43.5 -> 59.9Improved

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