NeurIPS 2020

Compositional Generalization via Neural-Symbolic Stack Machines


Meta Review

The paper proposes a new method for compositional generalization in sequence-to-sequence tasks. The basic idea is to have a symbolic stack machine (capable of compositionally manipulating sequences) that is controlled by a neural network. The method gets perfect accuracy on an existing compositional generalization dataset, a small-scale English-French machine translation task, and a grammar parsing task. The paper was well-received. Pros: + Novel architecture + Attractive way of providing inductive bias without hardcoding too much knowledge + The paper is well-written + Strong experimental results in the domains considered Cons: + The paper could do more by the way of providing insights about why the model works. The reviewers appreciated the clarifications provided in the author feedback. Please integrate these clarifications (in particular, the results of the ablation study and comparisons with differentiable data structures) into the main paper, and consult the reviews for more detailed feedback.