SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2009.08330 · EMNLP Findings · 2020

More Embeddings, Better Sequence Labelers?

Yong Jiang, Fei Huang, Tao Wang, Kewei Tu, Xinyu Wang, Nguyen Bach, Zhongqiang Huang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. We have no repository linked to it either. That is a statement about our graph, not about the paper.

Abstract

Recent work proposes a family of contextual embeddings that significantly improves the accuracy of sequence labelers over noncontextual embeddings. However, there is no definite conclusion on whether we can build better sequence labelers by combining different kinds of embeddings in various settings. In this paper, we conduct extensive experiments on 3 tasks over 18 datasets and 8 languages to study the accuracy of sequence labeling with various embedding concatenations and make three observations: (1) concatenating more embedding variants leads to better accuracy in rich-resource and cross-domain settings and some conditions of low-resource settings; (2) concatenating contextual sub-word embeddings with contextual character embeddings hurts the accuracy in extremely lowresource settings; (3) based on the conclusion of (1), concatenating additional similar contextual embeddings cannot lead to further improvements. We hope these conclusions can help people build stronger sequence labelers in various settings.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2009.08330")
get_code_for_paper("2009.08330")
have("2009.08330")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2009.08330.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2009.08330.svg)](https://syntology.ai/paper/2009.08330)

Connect an agent — have() is free.