# Question about NLQ task on clips

**URL:** <https://discuss.ego4d-data.org/t/question-about-nlq-task-on-clips/352>\
**Category:** Ego4D Challenges\
**Tags:** natural-language-queries\
**Created:** [July 5, 2023, 10:33pm UTC](https://discuss.ego4d-data.org/t/question-about-nlq-task-on-clips/352 "2023-07-05T22:33:28Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![davidyao99](https://avatars.discourse-cdn.com/v4/letter/d/2acd7d/32.png) [@davidyao99](https://discuss.ego4d-data.org/u/davidyao99)\
**Post date:** [July 5, 2023, 10:33pm UTC](https://discuss.ego4d-data.org/t/question-about-nlq-task-on-clips/352/1 "2023-07-05T22:33:28Z")

</div>

Just a sanity check, for the NLQ task, do we process the entire clip? I see that there is a clip\_start\_sec and clip\_end\_sec associated with every clip. Does this mean we might sometimes only use part of the clip for the NLQ queries? Thanks!