SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.05847 · NAACL · 2021

Targeted Adversarial Training for Natural Language Understanding

Jianfeng Gao, Hao Cheng, Xiaodong Liu, Hoifung Poon, Ichiro Kobayashi, Lis Pereira

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. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We present a simple yet effective Targeted Adversarial Training (TAT) algorithm to improve adversarial training for natural language understanding. The key idea is to introspect current mistakes and prioritize adversarial training steps to where the model errs the most. Experiments show that TAT can significantly improve accuracy over standard adversarial training on GLUE and attain new state-of-the-art zero-shot results on XNLI. Our code will be released at: https://github. com/namisan/mt-dnn. * Equal contribution. (a) BERT with standard fine-tuning (b) BERT with TAT fine-tuning

For agents

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

get_harvested_code_for_paper("2104.05847")
get_code_for_paper("2104.05847")
have("2104.05847")

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

curl https://syntology.ai/api/ran/2104.05847.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.05847.svg)](https://syntology.ai/paper/2104.05847)

Connect an agent — have() is free.