AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

BabyVLM: Data-Efficient Pretraining of VLMs Inspired by Infant Learning

Shengao Wang, Arjun Chandra, Aoming Liu, Venkatesh Saligrama, Boqing Gong

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 1 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 child-directed transferred data to SAYCam (ablation)BabyLLaVA-GPT2 · VTWTaccuracy: 0.6252 -> 0.6933Improved
add child-directed transferred data to SAYCam (ablation)BabyLLaVA-GPT2 · Labeled-Saccuracy: 0.4195 -> 0.5364Improved
add child-directed transferred data to SAYCam (ablation)CVCL · VTWTaccuracy: 0.6494 -> 0.7021Improved
add child-directed transferred data to SAYCam (ablation)CVCL · Baby Winoground negative contextgroup score: 0.2246 -> 0.4493Improved
add child-directed transferred data to SAYCam (ablation)CVCL · Baby Winoground positive contextgroup score: 0.5068 -> 0.4657Hurt
add child-directed transferred data to SAYCam (ablation)CVCL · Labeled-Saccuracy: 0.6086 -> 0.5805Hurt
use transferred data instead of random general-domain data as the augmentation (ablation)BabyLLaVA-GPT2 · VTWTaccuracy: 0.6553 -> 0.6933Improved
use transferred data instead of random general-domain data as the augmentation (ablation)BabyLLaVA-GPT2 · SAYCam CaptionMETEOR: 0.1778 -> 0.1592Hurt
use transferred data instead of random general-domain data as the augmentation (ablation)CVCL · Baby Winoground overallgroup score: 0.1068 -> 0.2027Improved
double the SAYCam data by relaxing the filter threshold (ablation)BabyLLaVA-GPT2 · VTWTaccuracy: 0.6252 -> 0.6383Improved
add SAYCam anchor data to transferred data (ablation)BabyLLaVA-GPT2 · VTWTaccuracy: 0.6239 -> 0.6933Improved
add image input to VTWT evaluation (ablation)BabyLLaVA · VTWTaccuracy: 0.5360 -> 0.6252Improved

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