AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Q-Bench+: A Benchmark for Multi-Modal Foundation Models on Low-Level Vision From Single Images to Pairs

Zicheng Zhang, Haoning Wu, Erli Zhang, Guangtao Zhai, Weisi Lin

TPAMI 2024 · Vision-language and multimodal › Multimodal LLMs · approach: Instruction-tuned multimodal assistants · 60 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
use softmax probability of good vs poor instead of argmax (ablation)LLaVA-v1 (Vicuna-13B) · KonIQ-10kSRCC: 0.038 -> 0.462Improved
use softmax probability of good vs poor instead of argmax (ablation)mPLUG-Owl (LLaMA-7B) · KonIQ-10kSRCC: 0.111 -> 0.409Improved
use softmax probability of good vs poor instead of argmax (ablation)Shikra (Vicuna-7B) · KonIQ-10kSRCC: 0.178 -> 0.314Improved
use softmax probability of good vs poor instead of argmax (ablation)InstructBLIP (Vicuna-7B) · CGIQA-6KSRCC: 0.141 -> 0.263Improved
add prompt ensemble (good+high+fine vs poor+low+bad) (ablation)Emu2-Chat · IQA averageSRCC: 0.593 -> 0.625Improved
add prompt ensemble (good+high vs poor+low) (ablation)Qwen-VL · IQA averageSRCC: 0.475 -> 0.495Improved
add prompt ensemble (good+high+fine vs poor+low+bad) (ablation)LLaVA-v1.5 (Vicuna-v1.5-7B) · IQA averageSRCC: 0.424 -> 0.439Improved
add prompt ensemble (good+fine vs poor+bad) (ablation)InternLM-XComposer-VL · IQA averageSRCC: 0.541 -> 0.543No clear effect
fine-tune mPLUG-Owl2 on single-image low-level data (Q-Instruct)mPLUG-Owl2 · perception-pair testoverall accuracy: 48.94% -> 53.15%Improved
fine-tune mPLUG-Owl2 on single-image low-level data (Q-Instruct)mPLUG-Owl2 · description-pairprecision score: 0.92 -> 0.79Hurt
go from single images to image pairsjunior-level human · perceptionoverall accuracy: 74.31% -> 80.12%Improved
go from single images to image pairsGPT-4V · perceptionoverall accuracy: 74.10% -> 78.07%Improved

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