Repository navigation
Conversation
Answer-aware cascade: qwen/qwen3-235b-a22b-2507 answers every query; a fine-tuned native Clef-Flash answer critic scores the answer and escalates to deepseek/deepseek-v4-pro below a frozen threshold of 0.9. Escalated rows bill both calls (see generated_result.cascade_accounting). Resubmission of RouteWorks#236 with the 8,400 scored rows only; the optional optimality rows are omitted because two of their code answers exhaust the evaluation runner's memory. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Owner
Author
|
/evaluate |
1 similar comment
Owner
Author
|
/evaluate |
Owner
Author
|
Superseded: GitHub keeps inherited workflows disabled on forks, so the verification run moved to wmlba/routerarena-verification#1 (same trees, same unchanged workflow). |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Unofficial verification run on the wmlba fork. This is not a leaderboard submission.
This PR runs RouterArena's own
pr-evaluation.ymlworkflow, unchanged, on a standard GitHububuntu-latestrunner.main@04c99fd, plus the one-file LiveCodeBench memory fix proposed upstream in Limit memory of LiveCodeBench test processes RouteWorks/RouterArena#239 (issue LiveCodeBench evaluation has no memory limit: unbounded allocations end PR evaluations with exit code 143 RouteWorks/RouterArena#238), branchfix-lcb-memory-limit@3a67287.9e9b80f), unmodified.There are two purposes:
compass-coreroutersubmission while Add Compass CoreRouter submission (compass-corerouter, resubmission of #236) RouteWorks/RouterArena#237 waits for the upstream fix.The official result remains the one produced on RouteWorks#237 after RouteWorks#239 (or an equivalent fix) is merged.