Researchers have introduced THESEUS, a new framework for multi-hop knowledge graph question answering (KGQA) that reframes the task as a question-conditioned graph navigation problem. This approach aims to make the reasoning process explicit by modeling an agent that traverses relations within a knowledge graph to find an answer. To support this, the team has augmented existing datasets like KINSHIP and MQuAKE, developed new evaluation protocols to assess path fidelity and robustness, and adapted existing KG completion agents to work with natural-language question embeddings. AI
IMPACT Enhances explainability in knowledge graph question answering systems by making reasoning paths explicit.
RANK_REASON The cluster contains a research paper detailing a new framework and methodology for a specific AI task. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- CatalyzeX
- DagsHub
- Hugging Face
- KINSHIP
- MINERVA
- MQuAKE
- MultiHopKG
- Multi-Hop Knowledge Graph Question Answering
- SQUIRE
- THESEUS
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
