[vlm] refactor: reuse standard encoders for packed samples - #116
Merged
kaimo455 merged 3 commits intoAug 20, 2026
Merged
Conversation
Collaborator
|
@claude review this PR |
kaimo455
requested changes
Aug 10, 2026
kaimo455
force-pushed
the
codex/refactor-packed-vlm-encoders
branch
from
August 20, 2026 05:52
fe93957 to
0265d98
Compare
kaimo455
approved these changes
Aug 20, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Reuse the standard phase-aware VQA and multi-mix QA encoders for offline-packed samples, remove duplicated packing-only encoders, and keep assistant end-token supervision in the shared SFT VQA preprocessing path.
Rationale
Packing should only unpack, encode, and concatenate samples; standard encoder reuse keeps packed and non-packed behavior aligned and respects subclass overrides.