List your company
ResearchFormerly WizWand

Multi-hop Question Answering on Musique fixed split (test)

39.84Exact Match (EM)Graph-AdaptR1
-1.59369.163219.9230.6768May 29, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Graph-AdaptR1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Graph-based2026.0539.8453.42
Graph-R1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Graph-based2026.0537.548.33
Search-AdaptR1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.052534.31
Search-R1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.0517.1821.45
R1-SearcherBackbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.0510.1616.63
StandardRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Chunk-based2026.059.3817.31
GraphRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.057.0315.14
R1Backbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=None2026.057.0314.53
HippoRAG2Backbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.056.2512.37
HyperGraphRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.056.2520.4
SFTBackbone=Qwen2.5-7B-Instruct, Interaction Type=Training, Knowledge Representation=Chunk-based2026.055.4710.02
LightRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.053.9114.39
NaiveGenerationBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=None2026.053.1311.45
PathRAGBackbone=GPT-4o-mini, Interaction Type=Prompt Engineering, Knowledge Representation=Graph-based2026.053.1311.49
StandardRAGBackbone=Qwen2.5-7B-Instruct, Interaction Type=Prompt Engineering, Knowledge Representation=Chunk-based2026.050.784.53
NaiveGenerationBackbone=Qwen2.5-7B-Instruct, Interaction Type=Prompt Engineering, Knowledge Representation=None2026.0504.06