gemini-4POST
MathArena adds ArXivMath, BrokenArXiv and ArXivLean benchmarks with Gemini 4 Argon results
MathArena has released three new math benchmarks, ArXivMath, BrokenArXiv and ArXivLean, along with evaluation results for Gemini 4 Argon.
Jasper Dekoninck
@j_dekoninckArchived 20 days agoOriginally posted 2026.10.16
Jasper Dekoninck @j_dekoninckOriginally posted 2026.10.16
New ArXivMath, BrokenArXiv, ArXivLean and results for Gemini 4 Argon are out on MathArena! Go check them out :)
This post quotes another post: x.com/timgmath/status/2111058044507594981
x.com/j_dekoninck/status/2111059060485435817Archived 2026.10.17 10:03