Internet user attention spans are notoriously short - wait, I haven't lost you, have I? News summaries are in demand, and now Microsoft researchers are showing that AI is up for that task and making steps to reaching that level of natural language processing, according to VentureBeat.
The scientists from Microsoft’s Cambridge based research team explained their AI framework capable of discerning relationships in “weakly structured” text, beating out NLP alternatives for text summarization tasks, in their paper, “Structured Neural Summarization“ published on Arxiv.org. Their team trained the model using articles from CNN and the Daily Mail with accompanying summary sentences.
“Summarization, the task of condensing a large and complex input into a smaller representation that retains the core semantics of the input, is a classical task for natural language processing systems,” the researchers wrote. “Automatic summarization requires a machine learning component to identify important entities and relationships between them, while ignoring redundancies and common concepts … However, while standard [models] theoretically have the ability to handle arbitrary long-distance relationships, in practice they often fail to correctly handle long texts and are easily distracted by simple noise.”
"The key insight, which we believe to be widely applicable, is that inductive biases induced by explicit relationship modeling are a simple way to boost the practical performance of existing deep learning systems."
The model has two components: an extended sequence encoder (which can predict the next characters of the target sequence given an input sentence) and a neural network.
“We are excited about this initial progress and look forward to deeper integration of mixed
sequence-graph modeling in a wide range of tasks across both formal and natural languages,” the researchers wrote. “The key insight, which we believe to be widely applicable, is that inductive biases induced by explicit relationship modeling are a simple way to boost the practical performance of existing deep learning systems.”