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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Despite the success of neural models in solving reasoning tasks, theircompositional generalization capabilities remain unclear. In this work, wepropose a new setting of the structured explanation generation task tofacilitate compositional reasoning research. Previous works found that symbolicmethods achieve superior compositionality by using pre-defined inference rulesfor iterative reasoning. But these approaches rely on brittle symbolictransfers and are restricted to well-defined tasks. Hence, we propose a dynamicmodularized reasoning model, MORSE, to improve the compositional generalizationof neural models. MORSE factorizes the inference process into a combination ofmodules, where each module represents a functional unit. Specifically, we adoptmodularized self-attention to dynamically select and route inputs to dedicatedheads, which specializes them to specific functions. We conduct experiments forincreasing lengths and shapes of reasoning trees on two benchmarks to testMORSE's compositional generalization abilities, and find it outperformscompetitive baselines. Model ablation and deeper analyses show theeffectiveness of dynamic reasoning modules and their generalization abilities.

Dynamic MOdularized Reasoning for Compositional Structured Explanation Generation