Through the Looking Glass: Common Sense Consistency Evaluation of Weird Images

Elisei Rykov, Kseniia, Kseniia, Anton Razzhigaev, Alexander Panchenko, Vasily Konovalov

Through the Looking Glass: Common Sense Consistency Evaluation of Weird Images: 29 upvotes on Hugging Face Daily Papers, #9 of 43 papers on 2025-05-20. Day-by-day upvote history.

Measuring how real images look is a complex task in artificial intelligence research. For example, an image of a boy with a vacuum cleaner in a desert violates common sense. We introduce a novel method, which we call Through the Looking Glass (TLG), to assess image common sense consistency using Large Vision-Language Models (LVLMs) and Transformer-based encoder. By leveraging LVLMs to extract atomic facts from these images, we obtain a mix of accurate facts. We proceed by fine-tuning a compact attention-pooling classifier over encoded atomic facts. Our TLG has achieved a new state-of-the-art performance on the WHOOPS! and WEIRD datasets while leveraging a compact fine-tuning component.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.