We propose to utilize two auxiliary tasks, Number Ranking (NR) and Importance Ranking (IR), to supervise the encoder to capture the different relations.
Aug 1, 2021 · Table-to-text generation aims at automatically generating natural text to help people conve- niently obtain salient information in tables.
This repository provides the implementation for our paper "Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation" ACL2021 main ...
To this end, we first propose a two-level hierarchical encoder with coarse-to-fine attention to handle the attribute-value structure of the tables. Furthermore, ...
Missing: Improving | Show results with:Improving
Jul 27, 2021 · Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation ; Institute of Electrical and Electronics Engineers (IEEE) · Nakano ...
May 17, 2023 · Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation. record by Liang Li • Improving Encoder by Auxiliary ...
Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation ... However, it is hard for a vanilla encoder to capture these. 1.
Oct 22, 2024 · ... Table-to-Text Generation: Learning Better Representation ... Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation.
HierarchicalEncoder + NR + IR. 25.30, 55.88, 17.96. Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation. 2021. RotoWire. This dataset ...
A two-level hierarchical encoder with coarse-to-fine attention to handle the attribute-value structure of the tables, and proposes 3 joint tasks.