List your company

General Question Answering on TriviaQA (EM, LLM)

66.4Exact Match (EM)DAC
28.33638.21848.157.982Jun 9, 2026

Evaluation results

MethodLinks
DACBackbone=Qwen3-8B, FT=LoRA2026.0666.477.8
Search-R1 (GRPO/OF)Backbone=Qwen3-8B, FT=Full2026.0666.177.1
Search-R1 (GRPO/O)Backbone=Qwen3-8B, FT=Full2026.066676.6
Search-R1 (GRPO/O)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.066272
Search-R1 (PPO/OF)Backbone=Qwen3-8B, FT=Full2026.0661.873.1
Search-R1 (PPO/O)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0661.773.1
DACBackbone=Qwen2.5-7B-Instruct, FT=LoRA2026.0661.273.5
Search-R1 (GRPO/OF)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0660.972.9
Search-R1 (PPO/OF)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0660.871.8
Search-R1 (PPO/O)Backbone=Qwen3-8B, FT=Full2026.0660.570.6
Naive RAGBackbone=Qwen2.5-7B-Instruct, FT=-2026.0658.971.4
Naive RAGBackbone=Qwen3-8B, FT=-2026.0658.570.3
IRCoTBackbone=Qwen3-8B, FT=-2026.0657.974.5
Search-o1Backbone=Qwen2.5-7B-Instruct, FT=-2026.0652.365.2
Search-o1Backbone=Qwen3-8B, FT=-2026.0649.972.1
CoTBackbone=Qwen2.5-7B-Instruct, FT=-2026.064756.2
CoTBackbone=Qwen3-8B, FT=-2026.0645.353.8
DirectBackbone=Qwen2.5-7B-Instruct, FT=-2026.0644.852.3
DirectBackbone=Qwen3-8B, FT=-2026.0642.452.5
IRCoTBackbone=Qwen2.5-7B-Instruct, FT=-2026.0629.859.4