List your company

Question Answering on HotpotQA (Acc, F1)

56.3Accuracy (Acc)RELOOP
12.93224.19135.4546.709Oct 23, 2025

Evaluation results

MethodLinks
RELOOPAggregation Metric=best2025.1056.358.6
RELOOPAggregation Metric=median2025.1055.457.1
HippoRAGTask Specialization=Table + text2025.1053.255.7
AdaptiveRAGTask Specialization=Graph task2025.1050.352.5
Think on Graph (ToG)Category=RAG-based2025.1043.144.7
Graph-constrained Reasoning (GcR)Category=RAG-based2025.1039.241.6
Falcon-H1-7B-InstructCategory=LLM-only (direct QA)2025.1021.1—
Qwen3-4B-Instruct-2507Category=LLM-only (direct QA)2025.1017.8—
Falcon3-10B-instructCategory=LLM-only (direct QA)2025.1016.5—
Llama-3.1-8B-InstructCategory=LLM-only (direct QA)2025.1014.6—