activity
20232026
most citedHumanity's Last Exam

23 citations · 23 across the 6 of their papers we have counts for

collaborators

6 papers

cs.AI2026

COMPOSITE-Stem

Kyle Waters, Lucas Nuzzi, Tadhg Looram +20

AI agents hold growing promise for accelerating scientific discovery; yet, a lack of frontier evaluations hinders adoption into real workflows. Expert-written benchmarks have prove…

math.GR2026

Quasihomomorphisms to real algebraic groups

Sami Douba, Francesco Fournier-Facio, Sam Hughes +1

A quasihomomorphism is a map that satisfies the homomorphism relation up to bounded error. Fujiwara and Kapovich proved a rigidity result for quasihomomorphisms taking values in di…

math.GR2025

Stability, approximable quotients, and higher property (T)

Francesco Fournier-Facio

We construct a wealth of groups that are finitely presented, Frobenius stable, have property (T), but are very far from having property (T). Our method also shows that property…

math.GR2025

Stable reflection length in Coxeter groups

Francesco Fournier-Facio, Marco Lotz, Timothée Marquis

We introduce stable reflection length in Coxeter groups, as a way to study the asymptotic behaviour of reflection length. This creates connections to other well-studied stable leng…

cs.LG2025★ 23 cited

Humanity's Last Exam

Long Phan, Alice Gatti, Ziwen Han +1144

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achi…

math.GR2023

Infinite simple characteristic quotients

Rémi Coulon, Francesco Fournier-Facio

We construct continuum many infinite, simple, characteristic quotients of non-abelian free groups, answering a 1978 question of James Wiegold. The method is very flexible, allowing…