List your company

Exact Match on HotPotQA (Multi-hop Question Answering)

36Exact Match (HotPotQA)ReAct
0.649.821928.18Apr 5, 2026

Evaluation results

MethodLinks
ReActModel=GPT-5.42026.0436
ReActModel=Claude Haiku 4.52026.0426.5
ReActModel=Claude Sonnet 4.62026.0416.7
PTRModel=GPT-5.42026.0412.2
PTRModel=GPT-4o-Mini2026.0412
ReActModel=GPT-4o-Mini2026.048
PTRModel=Claude Sonnet 4.62026.048
PTRModel=Claude Haiku 4.52026.042