List your company

Multi-hop Question Answering on MorehopQA (AUROC, AUARC)

0.6457AUROCMATU
0.5152840.5491420.5830.616858Apr 9, 2026

Evaluation results

MethodLinks
MATULLM=Qwen2.5-7B2026.040.64570.7029
MATULLM=Llama3.1-8B2026.040.62620.6493
SAUP-MultipleLLM=Qwen2.5-7B2026.040.62420.6914
P(true)LLM=Qwen2.5-7B2026.040.62050.6853
P(true)LLM=Llama3.1-8B2026.040.61580.6416
Eigv(Agre)-finalLLM=Llama3.1-8B2026.040.6110.6369
SAUP-SingleLLM=Qwen2.5-7B2026.040.60880.677
Eigv(Agre)-finalLLM=Qwen2.5-7B2026.040.60790.6741
SAUP-MultipleLLM=Llama3.1-8B2026.040.60550.6322
Eigv(Agre)-WholeLLM=Llama3.1-8B2026.040.60340.63
Eigv(Agre)-WholeLLM=Qwen2.5-7B2026.040.60210.6681
SAUP-SingleLLM=Llama3.1-8B2026.040.59180.6277
MATULLM=GPT-4o2026.040.56710.4336
P(true)LLM=GPT-4o2026.040.54550.4121
Eigv(Agre)-finalLLM=GPT-4o2026.040.52570.3992
Eigv(Agre)-WholeLLM=GPT-4o2026.040.52030.401