English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recently, quite a few novel neural architectures were derived to solve mathword problems by predicting expression trees. These architectures varied fromseq2seq models, including encoders leveraging graph relationships combined withtree decoders. These models achieve good performance on various MWPs datasetsbut perform poorly when applied to an adversarial challenge dataset, SVAMP. Wepresent a novel model MMTM that leverages multi-tasking and multi-decoderduring pre-training. It creates variant tasks by deriving labels usingpre-order, in-order and post-order traversal of expression trees, and usestask-specific decoders in a multi-tasking framework. We leverage transformerarchitectures with lower dimensionality and initialize weights from RoBERTamodel. MMTM model achieves better mathematical reasoning ability andgeneralisability, which we demonstrate by outperforming the best state of theart baseline models from Seq2Seq, GTS, and Graph2Tree with a relativeimprovement of 19.4% on an adversarial challenge dataset SVAMP.

MMTM: Multi-Tasking Multi-Decoder Transformer for Math Word Problems