# Revisiting Text-to-Image Evaluation with Gecko: On Metrics, Prompts, and Human Ratings

arXiv: https://arxiv.org/abs/2404.16820 · HTML: https://mlsanity.com/paper/2404.16820/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 60 of its 91 references that are in the map.
- First posted on arXiv 25 Apr 2024.

## Builds on (most cited first, 30 of 60)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [Scaling Autoregressive Models for Content-Rich Text-to-Image Generation](https://mlsanity.com/paper/2206.10789/) (2022)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [On Faithfulness and Factuality in Abstractive Summarization](https://mlsanity.com/paper/2005.00661/) (2020)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [COCO-Stuff: Thing and Stuff Classes in Context](https://mlsanity.com/paper/1612.03716/) (2016)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [PaLI: A Jointly-Scaled Multilingual Language-Image Model](https://mlsanity.com/paper/2209.06794/) (2022)
- [Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies](https://mlsanity.com/paper/2101.02235/) (2021)
- [Muse: Text-To-Image Generation via Masked Generative Transformers](https://mlsanity.com/paper/2301.00704/) (2023)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding](https://mlsanity.com/paper/2109.14084/) (2021)
- [Evaluating the Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/1910.12840/) (2019)
- [Phenaki: Variable Length Video Generation From Open Domain Textual Description](https://mlsanity.com/paper/2210.02399/) (2022)
- [Evaluating Text-to-Visual Generation with Image-to-Text Generation](https://mlsanity.com/paper/2404.01291/) (2024)
- [Lumiere: A Space-Time Diffusion Model for Video Generation](https://mlsanity.com/paper/2401.12945/) (2024)
- [TIFA: Accurate and Interpretable Text-to-Image Faithfulness Evaluation with Question Answering](https://mlsanity.com/paper/2303.11897/) (2023)
- [DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models](https://mlsanity.com/paper/2210.14896/) (2022)

## Built on by (most cited first, 30 of 44)

- [Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps](https://mlsanity.com/paper/2501.09732/) (2025)
- [OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generation](https://mlsanity.com/paper/2506.07977/) (2025)
- [TIIF-Bench: How Does Your T2I Model Follow Your Instructions?](https://mlsanity.com/paper/2506.02161/) (2025)
- [GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation](https://mlsanity.com/paper/2512.16853/) (2025)
- [EvalMuse-40K: A Reliable and Fine-Grained Benchmark with Comprehensive Human Annotations for Text-to-Image Generation Model Evaluation](https://mlsanity.com/paper/2412.18150/) (2024)
- [Proactive Agents for Multi-Turn Text-to-Image Generation Under Uncertainty](https://mlsanity.com/paper/2412.06771/) (2024)
- [EVALALIGN: Supervised Fine-Tuning Multimodal LLMs with Human-Aligned Data for Evaluating Text-to-Image Models](https://mlsanity.com/paper/2406.16562/) (2024)
- [MMIG-Bench: Towards Comprehensive and Explainable Evaluation of Multi-Modal Image Generation Models](https://mlsanity.com/paper/2505.19415/) (2025)
- [Eval3D: Interpretable and Fine-grained Evaluation for 3D Generation](https://mlsanity.com/paper/2504.18509/) (2025)
- [MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer](https://mlsanity.com/paper/2509.16197/) (2025)
- [Automatic Evaluation for Text-to-image Generation: Task-decomposed Framework, Distilled Training, and Meta-evaluation Benchmark](https://mlsanity.com/paper/2411.15488/) (2024)
- [Dynamic Classifier-Free Diffusion Guidance via Online Feedback](https://mlsanity.com/paper/2509.16131/) (2025)
- [Evaluating Numerical Reasoning in Text-to-Image Models](https://mlsanity.com/paper/2406.14774/) (2024)
- [Erasing More Than Intended? How Concept Erasure Degrades the Generation of Non-Target Concepts](https://mlsanity.com/paper/2501.09833/) (2025)
- [Text-to-Image Diffusion Models Cannot Count, and Prompt Refinement Cannot Help](https://mlsanity.com/paper/2503.06884/) (2025)
- [A Comprehensive Study of Decoder-Only LLMs for Text-to-Image Generation](https://mlsanity.com/paper/2506.08210/) (2025)
- [Analyzing Quality, Bias, and Performance in Text-to-Image Generative Models](https://mlsanity.com/paper/2407.00138/) (2024)
- [EvolveDirector: Approaching Advanced Text-to-Image Generation with Large Vision-Language Models](https://mlsanity.com/paper/2410.07133/) (2024)
- [NTIRE 2025 challenge on Text to Image Generation Model Quality Assessment](https://mlsanity.com/paper/2505.16314/) (2025)
- [AtelierEval: Agentic Evaluation of Humans & LLMs as Text-to-Image Prompters](https://mlsanity.com/paper/2605.22645/) (2026)
- [Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation](https://mlsanity.com/paper/2607.05382/) (2026)
- [FineGRAIN: Evaluating Failure Modes of Text-to-Image Models with Vision Language Model Judges](https://mlsanity.com/paper/2512.02161/) (2025)
- [PairBench: Are Vision-Language Models Reliable at Comparing What They See?](https://mlsanity.com/paper/2502.15210/) (2025)
- [DetailMaster: Can Your Text-to-Image Model Handle Long Prompts?](https://mlsanity.com/paper/2505.16915/) (2025)
- [KITTEN: A Knowledge-Intensive Evaluation of Image Generation on Visual Entities](https://mlsanity.com/paper/2410.11824/) (2024)
- [TextInVision: Text and Prompt Complexity Driven Visual Text Generation Benchmark](https://mlsanity.com/paper/2503.13730/) (2025)
- [GraPE: A Generate-Plan-Edit Framework for Compositional T2I Synthesis](https://mlsanity.com/paper/2412.06089/) (2024)
- [Greedy Growing Enables High-Resolution Pixel-Based Diffusion Models](https://mlsanity.com/paper/2405.16759/) (2024)
- [Instruction-augmented Multimodal Alignment for Image-Text and Element Matching](https://mlsanity.com/paper/2504.12018/) (2025)
- [CROC: Evaluating and Training T2I Metrics with Pseudo- and Human-Labeled Contrastive Robustness Checks](https://mlsanity.com/paper/2505.11314/) (2025)

## Abstract (arXiv)

While text-to-image (T2I) generative models have become ubiquitous, they do not necessarily generate images that align with a given prompt. While previous work has evaluated T2I alignment by proposing metrics, benchmarks, and templates for collecting human judgements, the quality of these components is not systematically measured. Human-rated prompt sets are generally small and the reliability of the ratings -- and thereby the prompt set used to compare models -- is not evaluated. We address this gap by performing an extensive study evaluating auto-eval metrics and human templates. We provide three main contributions: (1) We introduce a comprehensive skills-based benchmark that can discriminate models across different human templates. This skills-based benchmark categorises prompts into sub-skills, allowing a practitioner to pinpoint not only which skills are challenging, but at what level of complexity a skill becomes challenging. (2) We gather human ratings across four templates and four T2I models for a total of >100K annotations. This allows us to understand where differences arise due to inherent ambiguity in the prompt and where they arise due to differences in metric and model quality. (3) Finally, we introduce a new QA-based auto-eval metric that is better correlated with human ratings than existing metrics for our new dataset, across different human templates, and on TIFA160.