Belebele (EN)

Observations

• ten sam task co PL → kontrola regresji PL↔EN ↗
• metryka: accuracy MCQ (accuracy) ↗
• rola: regresja ↗
• typ zadania: qa-mcq ↗
• licencja: CC-BY-SA ↗
• rozmiar: 900 ↗
• kategoria: en-regression ↗
• dodano: 2026-06-11 ↗
• dodał: slayer ↗
• status: verified ↗
• źródło: https://huggingface.co/datasets/facebook/belebele ↗
• 2026-06-04 mean across 4 runs: {“Qwen3.5-9B”:94.9} ↗
• 2026-06-04 Qwen3.5-9B: 94.9 (accuracy (MCQ, exact letter) [EN regresja]) ↗

Referenced by

• MCQ (mentions)
• EVAL (mentions)
• reading-comprehension (mentions)
• belebele_en (mentions)
• belebele_en (defined-by)
• belebele (mentions)
• belebele (defined-by)
• regresja (mentions)
• Belebele (EN) (mentions)
• Belebele (EN) (defined-by)
• V4 Faza 0 — pomiar (likelihood) + kalibracja λ (cal-l030) (uses-benchmark)
• Belebele (EN) (uses-benchmark)

Local graph

Provenance