This repo has the data, video feature, and model checkpoints used in our CVPR'24 paper ["Grounded Question-Answering in Long Egocentric Videos"](https://arxiv.org/abs/2312.06505).