|TREND||DATASET||BEST METHOD||PAPER TITLE||PAPER||CODE||COMPARE|
We predict separate convolution kernels based solely on the current time-step in order to determine the importance of context elements.
#6 best model for Machine Translation on WMT2014 English-German
There has been much recent work on training neural attention models at the sequence-level using either reinforcement learning-style methods or by optimizing the beam.
#7 best model for Machine Translation on IWSLT2015 German-English
Neural sequence-to-sequence models have provided a viable new approach for abstractive text summarization (meaning they are not restricted to simply selecting and rearranging passages from the original text).
#4 best model for Abstractive Text Summarization on CNN / Daily Mail
As part of this survey, we also develop an open source library, namely Neural Abstractive Text Summarizer (NATS) toolkit, for the abstractive text summarization.
In this survey, we consider seq2seq problems from the RL point of view and provide a formulation combining the power of RL methods in decision-making with sequence-to-sequence models that enable remembering long-term memories.
This paper presents a new Unified pre-trained Language Model (UniLM) that can be fine-tuned for both natural language understanding and generation tasks.
SOTA for Text Summarization on GigaWord (using extra training data)
Inspired by how humans summarize long documents, we propose an accurate and fast summarization model that first selects salient sentences and then rewrites them abstractively (i. e., compresses and paraphrases) to generate a concise overall summary.
For abstractive summarization, we propose a new fine-tuning schedule which adopts different optimizers for the encoder and the decoder as a means of alleviating the mismatch between the two (the former is pretrained while the latter is not).
SOTA for Extractive Document Summarization on CNN / Daily Mail (using extra training data)
We introduce a neural network model with a novel intra-attention that attends over the input and continuously generated output separately, and a new training method that combines standard supervised word prediction and reinforcement learning (RL).