Navigating with Graph Representations for Fast and Scalable Decoding of Neural Language Models
This addresses a bottleneck in NLP for faster and more scalable decoding, though it is an incremental improvement on existing methods.
The paper tackles the high computational cost of neural language models' softmax layer by proposing Fast Graph Decoder (FGD), which reduces decoding time by an order of magnitude while maintaining close to baseline accuracy on translation and language modeling tasks.
Neural language models (NLMs) have recently gained a renewed interest by achieving state-of-the-art performance across many natural language processing (NLP) tasks. However, NLMs are very computationally demanding largely due to the computational cost of the softmax layer over a large vocabulary. We observe that, in decoding of many NLP tasks, only the probabilities of the top-K hypotheses need to be calculated preciously and K is often much smaller than the vocabulary size. This paper proposes a novel softmax layer approximation algorithm, called Fast Graph Decoder (FGD), which quickly identifies, for a given context, a set of K words that are most likely to occur according to a NLM. We demonstrate that FGD reduces the decoding time by an order of magnitude while attaining close to the full softmax baseline accuracy on neural machine translation and language modeling tasks. We also prove the theoretical guarantee on the softmax approximation quality.