Research map / Vision-language and multimodal / Multimodal LLMs
Benchmarking Multimodal Large Language Models Against Image Corruptions
Xinkuan Qiu, Meina Kan, Yongbin Zhou, Shiguang Shan
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 changed | Where | Measured | Effect |
|---|---|---|---|
| increase corruption severity to most severe level | GPT-4o · MLLM-IC | accuracy: 14.3% drop | Hurt |
| increase corruption severity to most severe level | DeepSeek-VL · MLLM-IC | accuracy: 5.8% -> 9.8% drop | Hurt |
| blur corruption versus geometric transforms | GPT-4o · MLLM-IC | accuracy: 68.6 -> 53.0 | Hurt |
| blur corruption versus geometric transforms | Gemini 2.0 · MLLM-IC | accuracy: 77.2 -> 64.3 | Hurt |
| compression versus geometric corruption | Qwen2.5-VL · MLLM-IC | accuracy: 75.0 -> 64.1 | Hurt |
| spatial perception versus global context sensing | Qwen2.5-VL · MLLM-IC | accuracy: 73.2 -> 64.3 | Hurt |
| knowledge reasoning versus logical reasoning | LLaVA-1.5 · MLLM-IC | accuracy: 66.1 -> 47.0 | Hurt |
| larger or proprietary model Gemini 2.0 versus LLaVA-1.5 | Gemini 2.0 · MLLM-IC | accuracy: 63.6 -> 76.2 | Improved |
Closest papers by meaning
- Res-Bench: Benchmarking the Robustness of Multimodal Large Language Models to Dynamic Resolution Input (AAAI 2026 · similarity 0.91)
- MLLMGuard: A Multi-dimensional Safety Evaluation Suite for Multimodal Large Language Models (NeurIPS 2024 · similarity 0.91)
- SEED-Bench: Benchmarking Multimodal Large Language Models (CVPR 2024 · similarity 0.90)
- Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding? (ICML 2026 · similarity 0.90)
- SDEval: Safety Dynamic Evaluation for Multimodal Large Language Models (AAAI 2026 · similarity 0.89)
- VLM-RobustBench: A Comprehensive Benchmark for Robustness of Vision-Language Models (ICML 2026 · similarity 0.89)
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.
