How skis meet snow be the difference between winning gold or silver. Breakthroughs, discoveries, and DIY tips sent six days a week. Every ski and snowboarding event at the 2026 Winter Olympics is won through a combination of sheer athleticism, quick thinking, creativity, and persistence. But like so many other sports, competitors know their choice of equipment can mean the difference earning the gold or silver medal. A ski is built for function over form, and manufacturers have spent decades adapting and honing their products to ensure wearers get the best results.

Participants take part in a skijoring race in Zab, Poland, on January 25, 2026. Breakthroughs, discoveries, and DIY tips sent six days a week. The high-adrenaline winter sport of skijoring, derived from the Norwegian word for "ski driving," takes so many forms that it even defies uniform pronunciation. "If you go to France, it's skijoering, pronounced SKEE-zhor-ing. In German, it's skijöring, pronounced SHEE-yuh-ring," says Loren Zhimanskova, founder of Skijor International and Skijor USA.

Action Quality Assessment (AQA) aims to evaluate and score sports actions, which has attracted widespread interest in recent years. Existing AQA methods primarily predict scores based on features extracted from the entire video, resulting in limited interpretability and reliability. Meanwhile, existing AQA datasets also lack fine-grained annotations for action scores, especially for deduction items and sub-score annotations. In this paper, we construct the first AQA dataset containing fine-grained sub-score and deduction annotations for aerial skiing, which will be released as a new benchmark. For the technical challenges, we propose a novel AQA method, named JudgeMind, which significantly enhances performance and reliability by simulating the judgment and scoring mindset of professional referees. Our method segments the input action video into different stages and scores each stage to enhance accuracy. Then, we propose a stage-aware feature enhancement and fusion module to boost the perception of stage-specific key regions and enhance the robustness to visual changes caused by frequent camera viewpoints switching. In addition, we propose a knowledge-based grade-aware decoder to incorporate possible deduction items as prior knowledge to predict more accurate and reliable scores. Experimental results demonstrate that our method achieves state-of-the-art performance.

Instruction tuning large language models (LLMs) using machine-generated instruction-following data has been shown to improve zero-shot capabilities on new tasks, but the idea is less explored in the multimodal field.

Additionally, to prevent degenerative cases, we employ a distillation-based regularization by using a pretrained model.