Benchmark RadarAI BENCHMARK PROFILE
P3B3
P3B3 is an expert-curated benchmark of conversational prompts for evaluating variety bias and controllability in LLMs across European and Brazilian Portuguese.
- Released
- 2026-06-15
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Evaluates regional variety bias in Portuguese LLMs, addressing underrepresentation and controllability gaps.
Motivation
As Large Language Models (LLMs) become embedded in everyday communication, capturing regional linguistic variation is essential for reliable and equitable language use.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.