List your company

Single-Hop Knowledge Graph Question Answering on Simple Questions

85.8Hits@1Prior FT SOTA
16.22434.28752.3570.413Oct 18, 2024Jan 1, 2025Mar 17, 2025Jun 1, 2025Aug 15, 2025Oct 29, 2025Jan 13, 2026

Evaluation results

MethodLinks
Prior FT SOTAClass=SL, External Knowledge=With external knowledge2024.1085.8
PoGClass=ICL, LLM=GPT-4, External Knowledge=With external knowledge2024.1084
PrivGemoType=Proposed, Brain LLM=GPT-3.5-Turbo, Hand LLM=Qwen3-32b2026.0183.8
PoG-EClass=ICL, LLM=GPT-4, External Knowledge=With external knowledge2024.1081.2
PoGClass=ICL, LLM=GPT-3.5-Turbo, External Knowledge=With external knowledge2024.1080.8
Logits-to-LogicBackbone LLM=LLaMA3.1-8b2025.1180.3
PrivGemoType=Proposed, Brain LLM=DeepSeek-V3, Hand LLM=Qwen3-32b2026.0180.2
PrivGemoType=Proposed, Brain LLM=GPT-4o-mini, Hand LLM=Qwen3-32b2026.0179.6
PoG-EClass=ICL, LLM=GPT-3.5-Turbo, External Knowledge=With external knowledge2024.1078.3
KG-CoTMethod Paradigm=Agentic Reasoning, Backbone LLM=ChatGPT2025.1177.8
ToG/ToG-RClass=ICL, LLM=GPT-4, External Knowledge=With external knowledge2024.1066.7
ToGType=KG-centric RAG, LLM=GPT-42026.0166.7
ToGMethod Paradigm=Agentic Reasoning, Backbone LLM=GPT42025.1166.7
ToG-RMethod Paradigm=Agentic Reasoning, Backbone LLM=GPT42025.1158.6
ToG/ToG-RClass=ICL, LLM=GPT-3.5-Turbo, External Knowledge=With external knowledge2024.1053.6
ToGType=KG-centric RAG, LLM=GPT-3.5-Turbo2026.0153.6
ToGMethod Paradigm=Agentic Reasoning, Backbone LLM=ChatGPT2025.1153.6
ToG-RMethod Paradigm=Agentic Reasoning, Backbone LLM=ChatGPT2025.1145.4
CoTLLM=GPT-3.5-Turbo, External Knowledge=Without external knowledge2024.1020.3
CoTType=LLM-only, LLM=GPT-3.5-Turbo2026.0120.3
CoTMethod Paradigm=LLMs Reasoning, Backbone LLM=ChatGPT2025.1120.3
IO promptLLM=GPT-3.5-Turbo, External Knowledge=Without external knowledge2024.1020
IO promptType=LLM-only, LLM=GPT-3.5-Turbo2026.0120
IO promptMethod Paradigm=LLMs Reasoning, Backbone LLM=ChatGPT2025.1120
SCLLM=GPT-3.5-Turbo, External Knowledge=Without external knowledge2024.1018.9
SCType=LLM-only, LLM=GPT-3.5-Turbo2026.0118.9
SCMethod Paradigm=LLMs Reasoning, Backbone LLM=ChatGPT2025.1118.9