PReFIL combines LSTM question embeddings with two levels of convolutional features via 1x1 convolutions and recurrent spatial aggregation, setting new state-of-the-art accuracy on FigureQA and DVQA.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Answering Questions about Data Visualizations using Efficient Bimodal Fusion
PReFIL combines LSTM question embeddings with two levels of convolutional features via 1x1 convolutions and recurrent spatial aggregation, setting new state-of-the-art accuracy on FigureQA and DVQA.