AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

VITAL: Vision-Encoder-centered Pre-training for LMMs in Visual Quality Assessment

Ziheng Jia, Linhan Cao, Jinliang Han, Zicheng Zhang, Jiaying Qian, Jiarui Wang, Zijian Chen, Guangtao Zhai, Xiongkuo Min

CVPR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Adapting VLMs for zero-shot classification · 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 PMOD (machine opinion distribution) prediction instead of CE on mean level (ablation)VITAL-Base-8B · KADIDSRCC: 0.602 -> 0.759Improved
add PMOD prediction (ablation)VITAL-Base-8B · LIVE-VQCSRCC: 0.725 -> 0.800Improved
add pairwise preference training (ablation)VITAL-Base-8B · KADIDSRCC: 0.725 -> 0.759Improved
add text generation tasks to pre-training (ablation)VITAL-Base-8B · YT-GamingSRCC: 0.712 -> 0.710No clear effect
add text generation tasks to pre-training (ablation)VITAL-Base-8B · KADIDSRCC: 0.743 -> 0.759Improved
vision-encoder-centered pre-training on machine-annotated data vs SFT on human labelsInternVL3-8B · video scoring avg (8 datasets)AVG SRCC/PLCC: 0.758 -> 0.820Improved
pair VITAL encoder with new decoder without any training (Zero) vs original instruct modelInternVL3-1B · video scoring avg (8 datasets)AVG SRCC/PLCC: 0.405 -> 0.611Improved
warm-up decoder with 4000 samples after transferVITAL-Zero-1B · video scoring avg (8 datasets)AVG SRCC/PLCC: 0.611 -> 0.808Improved
transfer decoder to 14B with warm-upVITAL-Base-8B · video scoring avg (8 datasets)AVG SRCC/PLCC: 0.820 -> 0.795Mixed
post-train with VITAL pre-training before SFTVITAL-Assistant-8B · QBench-video-test-singleoverall accuracy (%): 42.67% -> 62.94%Improved
use focal loss instead of CE in post-training (ablation)VITAL-Assistant-8B · QBench-video open-ended questions (316)output length and accuracyImproved
linear probe on VITAL encoder vs Simple-VQAVITAL Vision Encoder · LIVE-VQCSRCC: 0.749 -> 0.760Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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