ImagenWorld: Stress-Testing Image Generation Models with Explainable Human Evaluation on Open-ended Real-World Tasks

• URL: http://arxiv.org/abs/2603.27862v1
• Date: Sun, 29 Mar 2026 20:42:05 GMT
• Title: ImagenWorld: Stress-Testing Image Generation Models with Explainable Human Evaluation on Open-ended Real-World Tasks
• Abstract summary: ImagenWorld is a benchmark of 3.6K condition sets spanning six core tasks (generation and editing, with single or multiple references) and six topical domains (artworks, photorealistic images, information graphics, textual graphics, computer graphics, and screenshots)<n>The benchmark is supported by 20K fine-grained human annotations and an explainable evaluation schema that tags localized object-level and segment-level errors.
• Score: 33.143423584118516
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Advances in diffusion, autoregressive, and hybrid models have enabled high-quality image synthesis for tasks such as text-to-image, editing, and reference-guided composition. Yet, existing benchmarks remain limited, either focus on isolated tasks, cover only narrow domains, or provide opaque scores without explaining failure modes. We introduce \textbf{ImagenWorld}, a benchmark of 3.6K condition sets spanning six core tasks (generation and editing, with single or multiple references) and six topical domains (artworks, photorealistic images, information graphics, textual graphics, computer graphics, and screenshots). The benchmark is supported by 20K fine-grained human annotations and an explainable evaluation schema that tags localized object-level and segment-level errors, complementing automated VLM-based metrics. Our large-scale evaluation of 14 models yields several insights: (1) models typically struggle more in editing tasks than in generation tasks, especially in local edits. (2) models excel in artistic and photorealistic settings but struggle with symbolic and text-heavy domains such as screenshots and information graphics. (3) closed-source systems lead overall, while targeted data curation (e.g., Qwen-Image) narrows the gap in text-heavy cases. (4) modern VLM-based metrics achieve Kendall accuracies up to 0.79, approximating human ranking, but fall short of fine-grained, explainable error attribution. ImagenWorld provides both a rigorous benchmark and a diagnostic tool to advance robust image generation.

Related papers

• Charts Are Not Images: On the Challenges of Scientific Chart Editing [66.38730113476677]
  textitFigEdit is a benchmark for scientific figure editing comprising over 30,000 samples.<n>Our benchmark demonstrates the profound limitations of pixel-level manipulation.<n>By releasing textitFigEdit, we aim to enable systematic progress in structure-aware figure editing.
  arXiv  Detail & Related papers  (2025-11-30T06:13:48Z)
• UniREditBench: A Unified Reasoning-based Image Editing Benchmark [52.54256348710893]
  This work proposes UniREditBench, a unified benchmark for reasoning-based image editing evaluation.<n>It comprises 2,700 meticulously curated samples, covering both real- and game-world scenarios across 8 primary dimensions and 18 sub-dimensions.<n>We fine-tune Bagel on this dataset and develop UniREdit-Bagel, demonstrating substantial improvements in both in-domain and out-of-distribution settings.
  arXiv  Detail & Related papers  (2025-11-03T07:24:57Z)
• PoSh: Using Scene Graphs To Guide LLMs-as-a-Judge For Detailed Image Descriptions [55.95282725491425]
  PoSh is a metric for detailed image description that uses scene graphs as structured rubrics to guide LLMs-as-a-Judge.<n>PoSh is replicable, interpretable and a better proxy for human raters than existing metrics.<n>We show that PoSh achieves stronger correlations with the human judgments in DOCENT than the best open-weight alternatives.
  arXiv  Detail & Related papers  (2025-10-21T20:30:20Z)
• GIE-Bench: Towards Grounded Evaluation for Text-Guided Image Editing [60.66800567924348]
  We introduce a new benchmark designed to evaluate text-guided image editing models.<n>The benchmark includes over 1000 high-quality editing examples across 20 diverse content categories.<n>We conduct a large-scale study comparing GPT-Image-1 against several state-of-the-art editing models.
  arXiv  Detail & Related papers  (2025-05-16T17:55:54Z)
• Benchmarking Counterfactual Image Generation [22.573830532174956]
  Generative AI has revolutionised visual content editing, empowering users to effortlessly modify images and videos.<n>To perform realistic edits in domains such as natural image or medical imaging, modifications must respect causal relationships.<n>We present a comparison framework to thoroughly benchmark counterfactual image generation methods.
  arXiv  Detail & Related papers  (2024-03-29T16:58:13Z)
• Holistic Evaluation of Text-To-Image Models [153.47415461488097]
  We introduce a new benchmark, Holistic Evaluation of Text-to-Image Models (HEIM) We identify 12 aspects, including text-image alignment, image quality, aesthetics, originality, reasoning, knowledge, bias, toxicity, fairness, robustness, multilinguality, and efficiency. Our results reveal that no single model excels in all aspects, with different models demonstrating different strengths.
  arXiv  Detail & Related papers  (2023-11-07T19:00:56Z)
• ImagenHub: Standardizing the evaluation of conditional image generation models [48.51117156168]
  This paper proposes ImagenHub, which is a one-stop library to standardize the inference and evaluation of all conditional image generation models. We design two human evaluation scores, i.e. Semantic Consistency and Perceptual Quality, along with comprehensive guidelines to evaluate generated images. Our human evaluation achieves a high inter-worker agreement of Krippendorff's alpha on 76% models with a value higher than 0.4.
  arXiv  Detail & Related papers  (2023-10-02T19:41:42Z)
• Bridging Composite and Real: Towards End-to-end Deep Image Matting [88.79857806542006]
  We study the roles of semantics and details for image matting. We propose a novel Glance and Focus Matting network (GFM), which employs a shared encoder and two separate decoders. Comprehensive empirical studies have demonstrated that GFM outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-10-30T10:57:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.