Using SocialReasoning Bench, we observed a stable pattern across models—agents execute competently…
Microsoft Research
Visit on X@MSFTResearch555K followers
12,812 views
Using SocialReasoning Bench, we observed a stable pattern across models—agents execute competently, but fail to consistently improve the user’s position, even with explicit instructions to optimize for user interest.
Comments
Comments from the original post on X (Twitter)
No comments yet
