Video Large Language Models excel in question answering but lack transparency in their decision-making process, providing answers without visual evidence. To address this, researchers have proposed Evidence-Backed Video Question Answering, a method that aims to provide verifiable visual grounding for answers. Current explainability efforts rely on textual rationales or sparse bounding boxes, which are insufficient for capturing complex video dynamics such as occlusions and non-rigid deformations1. This new approach has significant implications for the field of video analysis, as it enables more accurate and trustworthy question answering. The ability to provide evidence-backed answers is crucial in high-stakes applications, such as security and surveillance, where the consequences of incorrect or unverifiable answers can be severe. Therefore, the development of Evidence-Backed Video Question Answering matters to practitioners, as it has the potential to increase the reliability and accountability of video analysis systems.