List your company

Question Answering on SimpleQA Spanish-pivot CS (test)

17.1ES-ora ScoreES-ora
6.97049.600212.2314.8598Jun 18, 2026

Evaluation results

MethodLinks
ES-oraBackbone=Qwen1.5-27B, Configuration=Pure-Spanish baseline2026.0617.1————
ES-oraBackbone=Qwen1.5-9B, Configuration=Pure-Spanish baseline2026.0612.39————
ES-oraBackbone=Llama3.1-8B, Configuration=Pure-Spanish baseline2026.0610.54————
ES-oraBackbone=Mistral-7B, Configuration=Pure-Spanish baseline2026.068————
ES-oraBackbone=Qwen1.5-4B, Configuration=Pure-Spanish baseline2026.067.36————
BaseBackbone=Llama3.1-8B, Configuration=Direct CS answering2026.06—7.06———
BaseBackbone=Mistral-7B, Configuration=Direct CS answering2026.06—3.46———
BaseBackbone=Qwen1.5-4B, Configuration=Direct CS answering2026.06—3.6———
BaseBackbone=Qwen1.5-9B, Configuration=Direct CS answering2026.06—8.03———
BaseBackbone=Qwen1.5-27B, Configuration=Direct CS answering2026.06—11.68———
CANVASBackbone=Llama3.1-8B, Anchor=Spanish2026.06——8.53—1.47
CANVASBackbone=Mistral-7B, Anchor=Spanish2026.06——3.61—0.15
CANVASBackbone=Qwen1.5-4B, Anchor=Spanish2026.06——4.35—0.75
CANVASBackbone=Qwen1.5-9B, Anchor=Spanish2026.06——8.88—0.85
CANVASBackbone=Qwen1.5-27B, Anchor=Spanish2026.06——11.81—0.13
TGT-oraBackbone=Llama3.1-8B, Configuration=Pure-target baseline2026.06———2.57—
TGT-oraBackbone=Mistral-7B, Configuration=Pure-target baseline2026.06———1.16—
TGT-oraBackbone=Qwen1.5-4B, Configuration=Pure-target baseline2026.06———2.17—
TGT-oraBackbone=Qwen1.5-9B, Configuration=Pure-target baseline2026.06———3.85—
TGT-oraBackbone=Qwen1.5-27B, Configuration=Pure-target baseline2026.06———9.17—