List your company

Multi-Hop QA on 2Wiki (Accuracy)

70.5AccuracySLEA-RL
14.02828.68943.3558.011Mar 18, 2026

Evaluation results

MethodLinks
SLEA-RLBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0370.5
IGPOBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0368.2
GSPOBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0360.1
PPOBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0359.7
GRPOBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0357.7
RLOOBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0355
Reinforce++Base Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0354.5
GiGPOBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0343.6
SkillRLBase Model=Qwen2.5-7B-Instruct, Method Category=RL Methods2026.0343.2
EvolveRBase Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0338.2
Search-R1Base Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0337
ZeroSearchBase Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0334.6
RAGBase Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0325.8
R1-InstructBase Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0320.8
Search-o1Base Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0317
Qwen2.5Base Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0316.4
CoTBase Model=Qwen2.5-7B-Instruct, Method Category=Prompting2026.0316.2