List your company

Multi-hop Question Answering on 2WikiMultiHopQA (SR)

78SRGemini-3-Flash
30.1642.585567.42May 13, 2026

Evaluation results

MethodLinks
Gemini-3-FlashProtocol=Prompting2026.0578
Gemini-2.5-flashProtocol=Prompting2026.0566
Qwen3-30B-A3BProtocol=Prompting2026.0558
ICRLBackbone=Qwen3-8B2026.0550
Critique-GRPOBackbone=Qwen3-8B2026.0547.5
MATPOBackbone=Qwen3-8B2026.0546
ICRLBackbone=Qwen3-4B2026.0545
GSPOBackbone=Qwen3-8B2026.0544.5
Critique-GRPOBackbone=Qwen3-4B2026.0543
MATPOBackbone=Qwen3-4B2026.0542
GRPOBackbone=Qwen3-8B2026.0542
ScalingInterBackbone=Qwen3-8B2026.0540
ScalingInterBackbone=Qwen3-4B2026.0538
GRPOBackbone=Qwen3-4B2026.0536
GSPOBackbone=Qwen3-4B2026.0536
Qwen3-4BProtocol=Prompting2026.0532
Qwen3-8BProtocol=Prompting2026.0532