AI Paper Atlas

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

OpenVision 2: A Family of Generative Pretrained Visual Encoders for Multimodal Learning

Yanqing Liu, Xianhang Li, Letian Zhang, Zirui Wang, Zeyu Zheng, Yuyin Zhou, Cihang Xie

CVPR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 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
remove text encoder and contrastive loss (caption-only training)OpenVision L/14 · Open-LLaVA-Next (336px)TextVQA / OCRBench: TextVQA 68.3 -> 68.9; OCR 547 -> 537No clear effect
remove text encoder and contrastive lossOpenVision L/14 · Open-LLaVA-Next (336px)ChartQA: 68.0 -> 62.3Hurt
remove text encoder and contrastive lossOpenVision L/14 · LLaVA-1.5 (336px)TextVQA / OCRBench: TextVQA 61.2 -> 63.0; OCR 339 -> 357Improved
remove text encoder and contrastive lossOpenVision L/14 · Training time (TPU v4-512)hours: 83 -> 57Improved
remove text encoder and contrastive lossOpenVision L/14 · Peak memory, batch 2kGB per chip: 24.5 -> 13.8Improved
add CLIPA low-res optimization and token masking (ablation)CapPa baseline · Training time L/14 224pxhours: 217 -> 55Improved
replace alt-text with synthetic ReCap captions (ablation)OpenVision 2 · Caption type ablationTextVQA / OCRBench: TextVQA 51.8 -> 56.5; OCR 238 -> 303Improved
use ReCap v2 (alt-text conditioned) instead of v1 captions (ablation)OpenVision 2 · Caption type ablationTextVQA / OCRBench: TextVQA 56.9 -> 56.5; OCR 291 -> 303Mixed
mask visual tokens (keep 35%) instead of keeping 100% (ablation)OpenVision 2 · Token keep ratio ablationTextVQA / OCRBench: TextVQA 53.8 -> 56.9; OCR 254 -> 291Improved
keep only 10% of tokens instead of 35% (ablation)OpenVision 2 · Token keep ratio ablationTextVQA / OCRBench: TextVQA 56.9 -> 55.6; OCR 291 -> 276Hurt
increase max caption length from 16 to 128 tokens (ablation)OpenVision 2 · Max caption length ablationTextVQA / OCRBench: TextVQA 54.8 -> 56.5; OCR 269 -> 303Improved
add auxiliary MAE pixel reconstruction loss (ablation)OpenVision 2 · multimodal benchmarksbenchmark scoresHurt

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