List your company

Question Answering on SCQA Q3

100AccuracyHelicase
1.226.8552.578.15May 26, 2026

Evaluation results

MethodLinks
HelicaseWeb Search capability=true2026.0510065
Claude Opus 4.6Web Search capability=false2026.05550
ReActBackbone=Qwen3-235B, Web Search capability=true2026.054545
ToTBackbone=Qwen3-235B, Web Search capability=true2026.053535
GLM-5Web Search capability=false2026.05200
DeepSeek-V3.2Web Search capability=false2026.05150
Qwen3-235BWeb Search capability=false2026.0550