
State-of-the-art models struggle with Spacewalk-18 tasks, highlighting the need for improved video-language models.


coremsg

spacewalk-18-a-benchmark-for-multimodal-and-long-form-procedural-video-understanding


Spacewalk-18: A Benchmark for Multimodal and Long-form Procedural Video Understanding


title_rewrite


Spacewalk-18 introduces a challenging benchmark for procedural video understanding, highlighting the difficulty of task recognition and segmentation in a multimodal and long-form context.


spacewalk-18-benchmark-for-multimodal-procedural-video-understanding


Spacewalk-18: Benchmark for Multimodal Procedural Video Understanding
