作者位次 #3/7ACL 2026Poster· Natural Language Processing
Macaron: Controlled, Human-Written Benchmark for Multilingual and Multicultural Reasoning via Template-Filling
Alaa Elsetohy · Sama Hadhoud · Haryo Akbarianto Wibowo · Chenxi Whitehouse · Genta Indra Winata · Fajri Koto · Alham Fikri Aji
摘要
Multilingual benchmarks rarely test reasoning over culturally grounded premises: translated datasets keep English-centric scenarios, while culture-first datasets often lack control over the reasoning required. We propose Macaron, a template-first benchmark that factorizes reasoning type and cultu...