Coke, proposed by Dong et al. (2024a), minimizes LLM calls for Knowledge Graph Question Answering by using cluster-level Thompson sampling to formulate accuracy expectations and an optimized context-aware policy to select expert models based on question semantics.

Authors

• Person: Not available Organization: arXiv
  Large Language Models Meet Knowledge Graphs for Question ...

Sources

• Large Language Models Meet Knowledge Graphs for Question ... arxiv.org arXiv via serper

Referenced by nodes (1)

• Knowledge Graph Question Answering concept