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

Hybrid question answering (HQA) aims to answer questions over heterogeneousdata, including tables and passages linked to table cells. The heterogeneousdata can provide different granularity evidence to HQA models, e.t., column,row, cell, and link. Conventional HQA models usually retrieve coarse- orfine-grained evidence to reason the answer. Through comparison, we find thatcoarse-grained evidence is easier to retrieve but contributes less to thereasoner, while fine-grained evidence is the opposite. To preserve theadvantage and eliminate the disadvantage of different granularity evidence, wepropose MuGER$^2$, a Multi-Granularity Evidence Retrieval and Reasoningapproach. In evidence retrieval, a unified retriever is designed to learn themulti-granularity evidence from the heterogeneous data. In answer reasoning, anevidence selector is proposed to navigate the fine-grained evidence for theanswer reader based on the learned multi-granularity evidence. Experimentresults on the HybridQA dataset show that MuGER$^2$ significantly boosts theHQA performance. Further ablation analysis verifies the effectiveness of boththe retrieval and reasoning designs.

MuGER$^2$: Multi-Granularity Evidence Retrieval and Reasoning for Hybrid Question Answering