From Examples to Rules: Neural Guided Rule Synthesis for Information Extraction
This addresses the problem of making information extraction more accessible and adaptable for users without linguistic or domain expertise, though it is incremental in combining existing synthesis techniques with transformers.
The paper tackles the challenge of creating maintainable rule-based information extraction systems by synthesizing rules from examples, using a transformer-guided enumerative search to reduce exploration steps. It achieves state-of-the-art performance in 1-shot relation classification and competitive results in 5-shot scenarios without domain-specific training.
While deep learning approaches to information extraction have had many successes, they can be difficult to augment or maintain as needs shift. Rule-based methods, on the other hand, can be more easily modified. However, crafting rules requires expertise in linguistics and the domain of interest, making it infeasible for most users. Here we attempt to combine the advantages of these two directions while mitigating their drawbacks. We adapt recent advances from the adjacent field of program synthesis to information extraction, synthesizing rules from provided examples. We use a transformer-based architecture to guide an enumerative search, and show that this reduces the number of steps that need to be explored before a rule is found. Further, we show that without training the synthesis algorithm on the specific domain, our synthesized rules achieve state-of-the-art performance on the 1-shot scenario of a task that focuses on few-shot learning for relation classification, and competitive performance in the 5-shot scenario.