List your company

General Question Answering on NQ (Natural Questions) (EM, LLM)

49.2Exact Match (EM)DAC
14.98423.86732.7541.633Jun 9, 2026

Evaluation results

MethodLinks
DACBackbone=Qwen3-8B, FT=LoRA2026.0649.262.8
Search-R1 (GRPO/O)Backbone=Qwen3-8B, FT=Full2026.0648.862.2
Search-R1 (GRPO/OF)Backbone=Qwen3-8B, FT=Full2026.0648.161.8
Search-R1 (PPO/OF)Backbone=Qwen3-8B, FT=Full2026.0645.158.5
Search-R1 (GRPO/O)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0643.957.2
Search-R1 (PPO/O)Backbone=Qwen3-8B, FT=Full2026.0643.456.5
DACBackbone=Qwen2.5-7B-Instruct, FT=LoRA2026.0641.959.4
Search-R1 (GRPO/OF)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0639.657.9
Search-R1 (PPO/O)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0639.255.9
Search-R1 (PPO/OF)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0637.654.6
Naive RAGBackbone=Qwen2.5-7B-Instruct, FT=-2026.0636.955.4
Naive RAGBackbone=Qwen3-8B, FT=-2026.0635.954.9
IRCoTBackbone=Qwen3-8B, FT=-2026.0635.657.9
Search-o1Backbone=Qwen3-8B, FT=-2026.0632.856.3
Search-o1Backbone=Qwen2.5-7B-Instruct, FT=-2026.0631.351.8
CoTBackbone=Qwen3-8B, FT=-2026.0619.936
CoTBackbone=Qwen2.5-7B-Instruct, FT=-2026.0619.333.7
DirectBackbone=Qwen2.5-7B-Instruct, FT=-2026.0618.230.5
IRCoTBackbone=Qwen2.5-7B-Instruct, FT=-2026.0616.646.8
DirectBackbone=Qwen3-8B, FT=-2026.0616.332.5