-
I'd like to know where your code is used for Consistency guided Knowledge Denoising. also I want to know how to use LLama to train for DocRE tasks as I noticed that your data format is the same as Doc…
-
1. 我使用现有的英文数据进行训练，在使用test数据结果预测时，结果中只有：
"title": ，
"h_idx": ,
"t_idx": ,
"r": "",
但是没有evidence或者sent id，正常应该是得有evidence吧。请赐教，谢谢！
 DocRED dataset split. But i can't access it. I don't know if it's my fault, I would appreciate it if you could answer it or maybe you can provide a download addr…
-
Hi,
Thanks for your enlightening work! I have one question regarding the evaluation metrics on DocRED:
since DocRED has a different format of data from other joint RE dataset, i.e. one entity E_i co…
-
### Update
See https://docs.nestjs.com/discover/companies
-
> If you are only interested in the raw data for the tasks, you can directly download sampled episodes here: https://drive.google.com/drive/folders/1PuJSJxqZP4ijxFSBZZ6Fmc0SgR2S8pYU?usp=sharing (test_…
-
"Hello, when I use your model to run the Fully-supervised setting, the score is normal in the first step. It's also normal when I use it for dev evaluation. However, the score is 0 when I use it for t…
-
你好，我在复现您实验时得到的分数和您论文上的分数有0.6-1分的差距，我的参数设置和您发布的完全一样。请问我需要注意里面的什么参数吗，或者说您能帮我看看这差距的问题出在哪里吗，非常感谢
-
尊敬的作者您好，感谢您杰出的工作，您的工作为ChatGPT在信息抽取中的应用做出了很大的贡献！
最近在阅读您代码的过程中我发现您在文档级关系抽取数据集（如DocRED、Re-DocRED、DWIE）等数据集上进行的实验可能与原始的DocRED等数据集的定义不一致。
具体而言，您在文档级关系抽取的prompt中预先给定了可能有关系的实体对，而在DocRED论文定义（[https://ac…
-
I found that the {}-gc.json file was missing in the experiment of the reproduction model on the RE-DOCRED dataset, I would like to ask you how to generate this intermediate file?