Meta's Muse Spark 1.1 scores 48.5 on RadLE 2.0 radiology benchmark
X posts around a new radiology benchmark cast Meta's model as a front-runner, while also stressing that human experts still score higher.
On X, Alexandr Wang said Meta's Muse Spark 1.1 is state of the art on the Radiologists Last Exam Handover Readiness Index, or RadLE-H, and is nearing human expert performance, citing a related DrDatta_AIIMS post. The claim driving the conversation is that Muse Spark 1.1 scored 48.5, compared with a cited human expert baseline of 52.0, on a benchmark presented in those posts as a test of when a radiology AI should handle a case versus defer to a specialist.
Combined views
87.4K
8 posts, first seen 9h ago