Informed Named Entity Recognition Decoding for Generative Language Models
This work addresses the gap in applying generative models to information extraction tasks like NER, offering a future-proof method that improves performance and reduces hallucinations, though it is incremental in adapting existing generative techniques to a specific domain.
The paper tackled the problem of named entity recognition (NER) not benefiting from recent generative language models by proposing iNERD, which treats NER as a generative process with informed decoding. The approach achieved remarkable results, especially in environments with unknown entity classes, as evaluated on eight datasets with five generative models.
Ever-larger language models with ever-increasing capabilities are by now well-established text processing tools. Alas, information extraction tasks such as named entity recognition are still largely unaffected by this progress as they are primarily based on the previous generation of encoder-only transformer models. Here, we propose a simple yet effective approach, Informed Named Entity Recognition Decoding (iNERD), which treats named entity recognition as a generative process. It leverages the language understanding capabilities of recent generative models in a future-proof manner and employs an informed decoding scheme incorporating the restricted nature of information extraction into open-ended text generation, improving performance and eliminating any risk of hallucinations. We coarse-tune our model on a merged named entity corpus to strengthen its performance, evaluate five generative language models on eight named entity recognition datasets, and achieve remarkable results, especially in an environment with an unknown entity class set, demonstrating the adaptability of the approach.