{
  "id": 5298,
  "url": "https://arxiv.org/abs/2604.24929v1",
  "title": "GAIA-v2-LILT: Multilingual Adaptation of Agent Benchmark beyond Translation",
  "summary": "Agent benchmarks remain largely English-centric, while their multilingual versions are often built with machine translation (MT) and limited post-editing. We argue that, for agentic tasks, this minimal workflow can easily break benchmark validity through query-answer misalignment or culturally off-target context. We propose a refined workflow for adapting English benchmarks into multiple languages with explicit functional alignment, cultural alignment, and difficulty calibration using both autom",
  "authors": "Yunsu Kim, Kaden Uhlig, Joern Wuebker",
  "category": "research",
  "topics": "safety-alignment,agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-27T19:11:21.000Z",
  "fetched_at": "2026-07-14T16:31:40.219Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/5298",
  "original_url": "https://arxiv.org/abs/2604.24929v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}