AI Paper Atlas

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

2.5 Years in Class: A Multimodal Textbook for Vision-Language Pretraining

Wenqi Zhang, Hang Zhang, Xin Li, Jiashuo Sun, Yongliang Shen, Weiming Lu, Deli Zhao, Yueting Zhuang, Lidong Bing

ICCV 2025 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 14 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 LLM refinement of raw ASR text (ablation)Multimodal Textbook · 7-benchmark 1-shot averageaccuracy: 26.2 -> 31.1Improved
add OCR text from keyframes (ablation)Multimodal Textbook · 7-benchmark 1-shot averageaccuracy: 28.8 -> 31.1Improved
use SSIM instead of pixel-level keyframe extraction (ablation)Multimodal Textbook · 7-benchmark 1-shot averageaccuracy: 22.1 -> 31.1Improved
use SSIM instead of CLIP-based keyframe extraction (ablation)Multimodal Textbook · 7-benchmark 1-shot averageaccuracy: 24.6 -> 31.1Improved
pretrain on Textbook-6.5M instead of OBELICSIdefics2-8B-base continual · MathVistaaccuracy: 27.6 -> 29.7Improved
pretrain on Textbook-6.5M instead of MMC4LLaVA-1.5-7B · ScienceQA-IMG 4-shotaccuracy: 11.6 -> 37.3Improved
pretrain on Textbook-6.5M instead of OBELICSLLaVA-1.5-7B · OKVQA 0-shotaccuracy: 13.0 -> 10.2Hurt
pretrain on Textbook-6.5M instead of OBELICSLLaVA-1.5-7B · OKVQA 4-shotaccuracy: 37.5 -> 39.9Improved
pretrain on Textbook-6.5M instead of OBELICSIdefics2-8B from scratch · OKVQAaccuracy: 10.5 -> 10.1No clear effect
pretrain on Textbook-6.5M instead of OBELICSIdefics2-8B-base · MathVista 1-shot cheat testaccuracy: 67.7 -> 94.1Improved
pretrain on Textbook-6.5M instead of OBELICS then SFTLLaVA-1.5-7B · MathVista after SFTaccuracy: 25.6 -> 28.7Improved
shuffle image order in the dataset (20%, 50%, 100%) (ablation)Textbook-6.5M · 7-benchmark averageaccuracyHurt

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