docs(routing): the judge never sees client instructions - #311
Merged
Conversation
Signed-off-by: Ryan Lempka <rlempka@nvidia.com>
ryan-lempka
changed the title
ryan-lempka
marked this pull request as ready for review
August 5, 2026 19:15
|
Caution Review failedAn error occurred during the review process. Please try again later. Comment |
ryan-lempka
enabled auto-merge (squash)
August 5, 2026 19:17
|
ayushag-nv
approved these changes
Aug 5, 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The
recent_turn_windowdocs claim the judge sees client system/developer instructions. It does not, and it never has — inbound decoders normalize that content intoLlmRequest::instructions, which the judge request never reads.The packaged classifier prompt agrees with the code, not the docs: it tells the judge it receives the task's opening instruction and latest user follow-up. Correcting the docs and the matching
trim_messagescomment.