AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Depicting Beyond Scores: Advancing Image Quality Assessment through Multi-modal Language Models

Zhiyuan You, Zheyuan Li, Jinjin Gu, Zhenfei Yin, Tianfan Xue, Chao Dong

ECCV 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 101 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 Task-2 brief templated comparison data to training (ablation)DepictQA · BAPPS Traditionalcomparison accuracy: 66.4 -> 79.7Improved
add Task-3 reasoning data to training (ablation)DepictQA · M-BAPPS reasoning valreasonable rate: N/A -> 41.0 (Task2 vs Task2+Task3)Improved
add Task-1 description data without content description data (ablation)DepictQA · BAPPS Traditionalcomparison accuracy: 79.7 -> 78.1Hurt
add Task-1 description data (ablation)DepictQA · M-BAPPS reasoning valreasonable rate: 41.0 -> 45.0Improved
add IQA-unrelated content description data as regularization (ablation)DepictQA · BAPPS Traditionalcomparison accuracy: 78.1 -> 80.0Improved
add content description data (ablation)DepictQA · M-BAPPS reasoning valreasonable rate: 45.0 -> 53.0Improved
use unique tag per image instead of unified tag (ablation)DepictQA · 50 reviewed reasoning responsesconfusion rate: 24% -> 12%Improved
use unique tag instead of image embedding or unique projector (ablation)DepictQA · M-BAPPS reasoning valreasonable rate: 26.0 / 40.0 -> 53.0Improved
use local feature map instead of single global CLIP feature (ablation)DepictQA · M-BAPPS reasoning valreasonable rate: 34.0 -> 53.0Improved
initialize from MLLM pre-trained on high-level perception tasks (ablation)DepictQA (Vicuna-v1.5) · BAPPS Traditionalcomparison accuracy: 80.0 -> 81.2Improved
vary LoRA rank from 12 to 32 (ablation)DepictQA · BAPPS averageaccuracy: 81.8 / 81.9 / 81.8 / 82.1No clear effect
use LoRA rank 8 (ablation)DepictQA · BAPPSaccuracy: failHurt

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