Show simple item record

dc.contributor.authorGlass, Michael
dc.contributor.authorCanim, Mustafa
dc.contributor.authorGliozzo, Alfio
dc.contributor.authorChemmengath, Saneem
dc.contributor.authorKumar, Vishwajeet
dc.contributor.authorChakravarti, Rishav
dc.contributor.authorSil, Avi
dc.contributor.authorPan, FeiFei
dc.contributor.authorBharadwaj, Samarth
dc.contributor.authorFauceglia, Nicolas Rodolfo
dc.date.accessioned2023-02-03T21:27:15Z
dc.date.available2023-02-03T21:27:15Z
dc.date.issued2021-04
dc.identifier.citationGlass, Michael, Mustafa Canim, Alfio Gliozzo, Saneem Chemmengath, Vishwajeet Kumar, Rishav Chakravarti, Avi Sil, Feifei Pan, Samarth Bharadwaj, and Nicolas Rodolfo Fauceglia. "Capturing Row and Column Semantics in Transformer Based Question Answering over Tables." arXiv preprint arXiv:2104.08303. https://arxiv.org/pdf/2104.08303.pdf. April 2021.en_US
dc.identifier.urihttps://arxiv.org/pdf/2104.08303.pdf
dc.identifier.urihttps://doi.org/10.48550/arXiv.2104.08303
dc.identifier.urihttps://arxiv.org/abs/2104.08303
dc.identifier.urihttps://hdl.handle.net/20.500.13015/6497
dc.description.abstractTransformer based architectures are recently used for the task of answering questions over tables. In order to improve the accuracy on this task, specialized pre-training techniques have been developed and applied on millions of open-domain web tables. In this paper, we propose two novel approaches demonstrating that one can achieve superior performance on table QA task without even using any of these specialized pre-training techniques. The first model, called RCI interaction, leverages a transformer based architecture that independently classifies rows and columns to identify relevant cells. While this model yields extremely high accuracy at finding cell values on recent benchmarks, a second model we propose, called RCI representation, provides a significant efficiency advantage for online QA systems over tables by materializing embeddings for existing tables. Experiments on recent benchmarks prove that the proposed methods can effectively locate cell values on tables (up to ~98% Hit@1 accuracy on WikiSQL lookup questions). Also, the interaction model outperforms the state-of-the-art transformer based approaches, pre-trained on very large table corpora (TAPAS and TaBERT), achieving ~3.4% and ~18.86% additional precision improvement on the standard WikiSQL benchmark.en_US
dc.publisherarXiven_US
dc.rightsAttribution-NonCommercial-NoDerivs 3.0 United States*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/us/*
dc.titleCapturing Row and Column Semantics in Transformer Based Question Answering over Tablesen_US
dc.typeArticleen_US


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record

Attribution-NonCommercial-NoDerivs 3.0 United States
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-NoDerivs 3.0 United States