OpenING: A Comprehensive Benchmark for Judging Open-ended Interleaved Image-Text Generation

TL;DR

OpenING is a new comprehensive benchmark with 5,400 instances across 56 tasks designed to evaluate interleaved image-text generation, along with IntJudge, a model for assessing these outputs, revealing significant room for improvement in current methods.

Contribution

The paper introduces OpenING, a large-scale, diverse benchmark for interleaved image-text generation, and presents IntJudge, an effective evaluation model surpassing GPT-based evaluators.

Findings

Current methods show substantial room for improvement.

IntJudge achieves 82.42% agreement with human judgments.

OpenING covers diverse real-world scenarios for robust evaluation.

Abstract

Multimodal Large Language Models (MLLMs) have made significant strides in visual understanding and generation tasks. However, generating interleaved image-text content remains a challenge, which requires integrated multimodal understanding and generation abilities. While the progress in unified models offers new solutions, existing benchmarks are insufficient for evaluating these methods due to limitations in data size and diversity. To bridge this gap, we introduce OpenING, a comprehensive benchmark comprising 5,400 high-quality human-annotated instances across 56 real-world tasks. OpenING covers diverse daily scenarios such as travel guide, design, and brainstorming, offering a robust platform for challenging interleaved generation methods. In addition, we present IntJudge, a judge model for evaluating open-ended multimodal generation methods. Trained with a novel data pipeline, our IntJudge achieves an agreement rate of 82.42% with human judgments, outperforming GPT-based evaluators by 11.34%. Extensive experiments on OpenING reveal that current interleaved generation methods still have substantial room for improvement. Key findings on interleaved image-text generation are further presented to guide the development of next-generation models.