[GH-ISSUE #92] Sequence packing #74

Open
opened 2026-10-26 21:11:27 +00:00 by bsakaguchi · 0 comments
Owner

Originally created by @Waneila on GitHub (Feb 21, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-VL2/issues/92

In the DeepSESK-VL2 paper, what does "Sequence packing" mean? Does it refer to concatenating multiple samples into a single sequence? If Sequence packing is not used, what approach would be adopted for sequences that are too short? Is it padding?

Originally created by @Waneila on GitHub (Feb 21, 2025). Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-VL2/issues/92 In the DeepSESK-VL2 paper, what does "Sequence packing" mean? Does it refer to concatenating multiple samples into a single sequence? If Sequence packing is not used, what approach would be adopted for sequences that are too short? Is it padding?
Sign in to join this conversation.
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepSeek-VL2#74
No description provided.