LLMAEL: Large Language Models are Good Context Augmenters for Entity Linking
This addresses the challenge of disambiguating rare entities in entity linking, which is incremental as it builds on existing specialized models by augmenting them with LLM-generated context.
The paper tackles the problem of entity linking for long-tail entities by introducing LLMAEL, a framework that uses large language models as context augmenters to generate entity descriptions for specialized entity linking models, achieving state-of-the-art results with an 8.9% absolute gain in accuracy across 6 benchmarks.
Specialized entity linking (EL) models are well-trained at mapping mentions to unique knowledge base (KB) entities according to a given context. However, specialized EL models struggle to disambiguate long-tail entities due to their limited training data. Meanwhile, extensively pre-trained large language models (LLMs) possess broader knowledge of uncommon entities. Yet, with a lack of specialized EL training, LLMs frequently fail to generate accurate KB entity names, limiting their standalone effectiveness in EL. With the observation that LLMs are more adept at context generation instead of EL execution, we introduce LLM-Augmented Entity Linking (LLMAEL), the first framework to enhance specialized EL models with LLM data augmentation. LLMAEL leverages off-the-shelf, tuning-free LLMs as context augmenters, generating entity descriptions to serve as additional input for specialized EL models. Experiments show that LLMAEL sets new state-of-the-art results across 6 widely adopted EL benchmarks: compared to prior methods that integrate tuning-free LLMs into EL, LLMAEL achieves an absolute 8.9% gain in EL accuracy. We release our code and datasets.