Abstract:
Entity linking is a crucial step in many knowledge-intensive tasks.Traditional methods mainly rely on co-occurrence patterns, neglecting semantic information.To address this issue, we propose an entity linking agent for Question Answering, which leverages the reasoning and semantic understanding capabilities of large language models.The agent can actively search entity information and select relevant ones according to context.The effectiveness of agent has been verified on the 2WikiMultiHopQA dataset.
Chat is not available.