List your company

Multi-Hop Question Answering on HotpotQA (EM, LLM)

45.7EMDAC
17.124.52531.9539.375Jun 9, 2026

Evaluation results

MethodLinks
DACBackbone=Qwen3-8B, FT=LoRA2026.0645.762.8
Search-R1 (GRPO/OF)Backbone=Qwen3-8B, FT=Full2026.064562.3
Search-R1 (GRPO/O)Backbone=Qwen3-8B, FT=Full2026.0644.861
Search-R1 (PPO/OF)Backbone=Qwen3-8B, FT=Full2026.0642.258.3
DACBackbone=Qwen2.5-7B-Instruct, FT=LoRA2026.0640.357
Search-R1 (PPO/O)Backbone=Qwen3-8B, FT=Full2026.0640.355.6
Search-R1 (GRPO/O)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0638.554.3
Search-R1 (PPO/O)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.063854
Search-R1 (GRPO/OF)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0637.153
IRCoTBackbone=Qwen3-8B, FT=-2026.0636.953.5
Search-R1 (PPO/OF)Backbone=Qwen2.5-7B-Instruct, FT=Full2026.0635.551.2
Search-o1Backbone=Qwen3-8B, FT=-2026.0629.951.3
Naive RAGBackbone=Qwen2.5-7B-Instruct, FT=-2026.0629.342.7
Search-o1Backbone=Qwen2.5-7B-Instruct, FT=-2026.062944.8
Naive RAGBackbone=Qwen3-8B, FT=-2026.062942.9
CoTBackbone=Qwen3-8B, FT=-2026.0621.331.3
CoTBackbone=Qwen2.5-7B-Instruct, FT=-2026.0620.830.5
DirectBackbone=Qwen2.5-7B-Instruct, FT=-2026.0619.128
IRCoTBackbone=Qwen2.5-7B-Instruct, FT=-2026.0618.740.9
DirectBackbone=Qwen3-8B, FT=-2026.0618.228.1