{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/vlue-a-multi-task-benchmark-for-evaluating","title":"VLUE: A Multi-Task Benchmark for Evaluating Vision-Language Models","arxiv_id":"2205.15237","date":"2022-05-30","proceeding":null,"authors":["Wangchunshu Zhou","Yan Zeng","Shizhe Diao","Xinsong Zhang"],"abstract":"Recent advances in vision-language pre-training (VLP) have demonstrated impressive performance in a range of vision-language (VL) tasks. However, there exist several challenges for measuring the community's progress in building general multi-modal intelligence. First, most of the downstream VL datasets are annotated using raw images that are already seen during pre-training, which may result in an overestimation of current VLP models' generalization ability. Second, recent VLP work mainly focuses on absolute performance but overlooks the efficiency-performance trade-off, which is also an important indicator for measuring progress. To this end, we introduce the Vision-Language Understanding Evaluation (VLUE) benchmark, a multi-task multi-dimension benchmark for evaluating the generalization capabilities and the efficiency-performance trade-off (``Pareto SOTA'') of VLP models. We demonstrate that there is a sizable generalization gap for all VLP models when testing on out-of-distribution test sets annotated on images from a more diverse distribution that spreads across cultures. Moreover, we find that measuring the efficiency-performance trade-off of VLP models leads to complementary insights for several design choices of VLP. We release the VLUE benchmark to promote research on building vision-language models that generalize well to more diverse images and concepts unseen during pre-training, and are practical in terms of efficiency-performance trade-off.","url_abs":"https://arxiv.org/abs/2205.15237v1","url_pdf":"https://arxiv.org/pdf/2205.15237v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"vlue-a-multi-task-benchmark-for-evaluating","repo_url":"https://github.com/michaelzhouwang/vlue","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"vietnamese-natural-language-understanding","task_name":"Vietnamese Natural Language Understanding"},{"task_slug":"xlm-r","task_name":"XLM-R"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2205.15237","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}