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

Math Word Problem (MWP) solving needs to discover the quantitativerelationships over natural language narratives. Recent work shows that existingmodels memorize procedures from context and rely on shallow heuristics to solveMWPs. In this paper, we look at this issue and argue that the cause is a lackof overall understanding of MWP patterns. We first investigate how a neuralnetwork understands patterns only from semantics, and observe that, if theprototype equations are the same, most problems get closer representations andthose representations apart from them or close to other prototypes tend toproduce wrong solutions. Inspired by it, we propose a contrastive learningapproach, where the neural network perceives the divergence of patterns. Wecollect contrastive examples by converting the prototype equation into a treeand seeking similar tree structures. The solving model is trained with anauxiliary objective on the collected examples, resulting in the representationsof problems with similar prototypes being pulled closer. We conduct experimentson the Chinese dataset Math23k and the English dataset MathQA. Our methodgreatly improves the performance in monolingual and multilingual settings.

Seeking Patterns, Not just Memorizing Procedures: Contrastive Learning for Solving Math Word Problems