AIBench: Evaluating Visual-Logical Consistency in Academic Illustration Generation
This addresses the need for reliable evaluation of AI-generated academic illustrations, which is an incremental step in improving visual content generation for research papers.
The paper tackles the problem of evaluating whether AI-generated academic illustrations are logically consistent with the paper text, proposing AIBench, a benchmark using VQA for logic correctness and VLMs for aesthetics, and finds that performance gaps between models are larger than in general tasks, with test-time scaling boosting results.
Although image generation has boosted various applications via its rapid evolution, whether the state-of-the-art models are able to produce ready-to-use academic illustrations for papers is still largely unexplored. Directly comparing or evaluating the illustration with VLM is native but requires oracle multi-modal understanding ability, which is unreliable for long and complex texts and illustrations. To address this, we propose AIBench, the first benchmark using VQA for evaluating logic correctness of the academic illustrations and VLMs for assessing aesthetics. In detail, we designed four levels of questions proposed from a logic diagram summarized from the method part of the paper, which query whether the generated illustration aligns with the paper on different scales. Our VQA-based approach raises more accurate and detailed evaluations on visual-logical consistency while relying less on the ability of the judger VLM. With our high-quality AIBench, we conduct extensive experiments and conclude that the performance gap between models on this task is significantly larger than general ones, reflecting their various complex reasoning and high-density generation ability. Further, the logic and aesthetics are hard to optimize simultaneously as in handcrafted illustrations. Additional experiments further state that test-time scaling on both abilities significantly boosts the performance on this task.