Handle ReAct tool selection failures more gracefully #8114
Merged
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Outdated suggestions cannot be applied.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.
We've accumulated a few reports, sometimes with Google models (Gemma, Gemini) but also with Qwen, that when the model wants to
finish
, it fails to format that properly. This results in an unnecessary exception, instead of ending execution gracefully.This PR refines the base instructions for
dspy.ReAct
and handles mid-trajectory failures more gracefully. The final extraction step is still responsible for processing. This results in better respect for the ReAct contract.