AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

LLMScore: Unveiling the Power of Large Language Models in Text-to-Image Synthesis Evaluation

Yujie Lu, Xianjun Yang, Xiujun Li, Xin Eric Wang, William Yang Wang

NeurIPS 2023 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 8 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 LLM-based instruction-following evaluation with multi-granularity visual descriptions rather than text-image matchingText-to-image alignment evaluation · Concept Conjunction and Attribute Binding ContrastKendall's tau: 0.4058 -> 0.4871; 0.3768 -> 0.5167; 0.1982 -> 0.4005; 0.1500 -> 0.3955Improved
Use LLM-based instruction-following evaluation with multi-granularity visual descriptions rather than text-image matchingText-to-image alignment evaluation · COCO2014, COCO2017, DrawBench, PaintSkillsKendall's tau: 0.3465 -> 0.3629; 0.2793 -> 0.3357; 0.1569 -> 0.2230; 0.4911 -> 0.5600Improved
Use GPT-4 rather than GPT-3.5 as evaluator (ablation)LLMScore · GeneralBench, OverallKendall's tau: 0.2480 -> 0.2793Improved
Use GPT-4 rather than GPT-3.5 as evaluator (ablation)LLMScore · GeneralBench, Error CountingKendall's tau: 0.2205 -> 0.2131Hurt
Replace LLM reasoning with caption-to-prompt matching (ablation)LLMScore evaluation pipeline · GeneralBenchKendall's tauHurt
Use visual-description-to-prompt matching instead of LLM evaluation (ablation)LLMScore evaluation pipeline · GeneralBenchKendall's tauHurt
Use GPT-4 visual descriptions with sentence-matching metrics (ablation)DescCLIP and DescMETEOR · GeneralBenchKendall's tau: DescCLIP Overall: 0.1479 -> 0.1128; DescMETEOR Overall: 0.0042 -> 0.0297; DescCLIP Error Counting: 0.0467 -> 0.0149; DesMixed

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