List your company

Question Answering on HotpotQA (test) (EM Metric Only)

42.5EMSD-Search-Base
13.79621.24828.736.152May 18, 2026

Evaluation results

MethodLinks
SD-Search-BaseTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0542.5
SD-Search-InstructTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0542.4
MR-Search-BaseTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0541.9
AutoRefine-BaseTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0540.5
AutoRefine-InstructTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0540.4
Thinker-InstructTraining Pipeline=Synthesize → SFT, External Teacher=Qwen2.5-72B, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0540
ReSearch-InstructTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0535.1
StepSearch-InstructTraining Pipeline=Pure RL, External Teacher=GPT-4o, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0534.5
Search-R1-InstructTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0533.1
StepSearch-BaseTraining Pipeline=Pure RL, External Teacher=GPT-4o, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0532.9
ReSearch-BaseTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0530.5
Search-R1-BaseTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0529.7
Naive RAGTraining Pipeline=Zero-Shot, External Teacher=–, Retrieval Strategy=w/ Single-Hop Retrieval, Backbone=Qwen2.5-3B2026.0525.5
Search-o1Training Pipeline=Prompting, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0522.1
R1-InstructTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/o Retrieval, Backbone=Qwen2.5-3B2026.0520.8
R1-BaseTraining Pipeline=Pure RL, External Teacher=–, Retrieval Strategy=w/o Retrieval, Backbone=Qwen2.5-3B2026.0520.1
SFTTraining Pipeline=SFT, External Teacher=–, Retrieval Strategy=w/o Retrieval, Backbone=Qwen2.5-3B2026.0518.6
IRCoTTraining Pipeline=Prompting, External Teacher=–, Retrieval Strategy=w/ Multi-Hop Retrieval, Backbone=Qwen2.5-3B2026.0516.4
Direct GenerationTraining Pipeline=Zero-Shot, External Teacher=–, Retrieval Strategy=w/o Retrieval, Backbone=Qwen2.5-3B2026.0514.9