-
* focus on a single job in workflow, collect multiple samples from different runs
* figure out hard-to-learn (low confidence, low variability) and ambiguous (low confidence, high variability)
re…
-
# Adversarial Transfer Learning for Chinese Named Entity Recognition with Self-Attention Mechanism
- authors: Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao, Shengping Liu
- link: https://www.aclweb.org/…
himkt updated
5 years ago
-
## 简介
non-autogression decoding. 用iterative的方式不断refine已有的翻译结果。第一轮是完全NAT,然后再次基础上,选择confidence差的N个词,mask掉,去对mask的内容进行refine,迭代X轮结束。对mask的恢复过程类似BERT。整篇文章读起来很舒服,听起来也合理。后面实验部分也很饱满。
## 论文信息
* Author: F…
-
**Summary:**
使用DL based的NER需要有大量标注数据,但人工标注很贵,所以使用distant supervision,但是这种方法产生了很多noise。这个工作提出了两个模型,针对一个典型的sequence labeling 任务,提出了一个fuzzy CRF layer来hadle token with multiple possible labels。另外提出一个了T…
-
- [x] Email template
- [x] The account email points to https://virtual.emnlp.2020.org which should be https://virtual.2020.emnlp.org instead.
- [x] The login instruction doc is [here](https://do…
-
https://aclanthology.org/D17-1238/
-
If we can distill or prune the NLLB-200 shortly after starting fine tuning, we may be able to dramatically reduce (50% or more) the training and inferencing time needed. It could even do something lik…
-
**Please be sure to add the Anthology ID in the title**
[See here](https://www.aclweb.org/anthology/info/corrections/) to read about the three types of corrections.
## Revision or erratum: pleas…
yg211 updated
3 years ago
-
What if two events starts at the same time?
How to show them one beside another?
Something like this https://virtual.2020.emnlp.org/schedule.html
Codesandbox
https://codesandbox.io/s/chakra-ui-c…
-
Hi Eric,
This is really cool work.
Do you have any idea on when you are going to release your code? We need to perform some experiments on your paper as a baseline for our EMNLP submission.
…