AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Information Density Principle for MLLM Benchmarks

Chunyi Li, Xiaozhe Li, Zicheng Zhang, Yuan Tian, Ziheng Jia, Xiaohong Liu, Xiongkuo Min, Jia Wang, Haodong Duan, Kai Chen, Guangtao Zhai

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Benchmarks and analysis of MLLMs · 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
average over time: newer benchmarks raise difficulty and cut fallacyMLLM benchmarks · 19 MLLM benchmarksDifficulty and FallacyImproved
newer benchmarks over time on diversityMLLM benchmarks · 19 MLLM benchmarksDiversityNo clear effect
making a benchmark harder (MME-Realworld) raises fallacy and lowers diversityMME-Realworld · MME-RealworldFallacy / Diversity: Difficulty 0.666, Fallacy 0.480, Diversity 0.701Mixed
Model Eval versus Data Eval for estimating human judgments (ablation)Information density pipeline · 19 MLLM benchmarkscorrelation (SRCC/PLCC)Mixed
automated evaluation agreement with human annotation (ablation)Information density pipeline · 19 MLLM benchmarkscorrelation (SRCC/PLCC): above 0.7Improved

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