Fix MiniMax-M3 reasoning-token handling in validators#17
Closed
ywang96 wants to merge 1 commit into
Closed
Conversation
ywang96
force-pushed
the
fix/minimax-m3-mm-think-support
branch
2 times, most recently
from
ba85185 to
0c9568e
Compare
MiniMax-M3 wraps its chain-of-thought in a native <mm:think>...</mm:think> token (both <think> and <mm:think> are special tokens in the M3 tokenizer), carried inline in `content`. ScenarioCheckValidator._get_visible_content only stripped <think>...</think>, so the parameter-order check read the order from inside the reasoning instead of the visible answer, causing spurious Scenario-Check-Pass-Rate failures on M3. Strip both tags. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ywang96
force-pushed
the
fix/minimax-m3-mm-think-support
branch
from
0c9568e to
8936e41
Compare
ywang96
changed the title
<mm:think>) handling in validators<mm:think> reasoning token in Scenario-Check
ywang96
changed the title
<mm:think> reasoning token in Scenario-Check
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
MiniMax-M3 wraps its chain-of-thought in a native
<mm:think>...</mm:think>token (both<think>and<mm:think>are special tokens in the M3 tokenizer), carried inline in the messagecontent.Fix
Strip both tags:
One-line change in
validator/scenario_check.py; behavior for<think>is unchanged.🤖 Generated with Claude Code