Cell-aware Stacked LSTMs for Modeling Sentences
This work addresses sentence modeling for natural language processing applications, representing an incremental improvement over existing stacked LSTM architectures.
The paper tackled the problem of modeling sentences by proposing a Cell-aware Stacked LSTM (CAS-LSTM) that uses both hidden and memory cell states from lower layers, resulting in significant performance gains over standard LSTMs on benchmark datasets for tasks like natural language inference and sentiment classification.
We propose a method of stacking multiple long short-term memory (LSTM) layers for modeling sentences. In contrast to the conventional stacked LSTMs where only hidden states are fed as input to the next layer, the suggested architecture accepts both hidden and memory cell states of the preceding layer and fuses information from the left and the lower context using the soft gating mechanism of LSTMs. Thus the architecture modulates the amount of information to be delivered not only in horizontal recurrence but also in vertical connections, from which useful features extracted from lower layers are effectively conveyed to upper layers. We dub this architecture Cell-aware Stacked LSTM (CAS-LSTM) and show from experiments that our models bring significant performance gain over the standard LSTMs on benchmark datasets for natural language inference, paraphrase detection, sentiment classification, and machine translation. We also conduct extensive qualitative analysis to understand the internal behavior of the suggested approach.