List your company

Multi-hop Reasoning on LongBench MuSiQue and WikiMultiHopQA

69.9F1 ScoreMGRS
57.10860.42963.7567.071Nov 28, 2025

Evaluation results

MethodLinks
MGRSBackbone=GPT-4o-mini2025.1169.9
AoTBackbone=GPT-4o-mini2025.1168.5
AFlowBackbone=GPT-4o-mini2025.1161
FoTBackbone=GPT-4o-mini, n=82025.1159.1
CoT-SCBackbone=GPT-4o-mini, n=52025.1158.6
Self-refineBackbone=GPT-4o-mini2025.1158.2
CoTBackbone=GPT-4o-mini2025.1157.6