Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
pandas
License:
sheryc commited on
Commit
c005691
·
verified ·
1 Parent(s): fbbcf07

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -53,6 +53,7 @@ language:
53
  <a href='https://github.com/FoundationAgents/CARE'><img src='https://img.shields.io/badge/_Code-CARE-181717?color=121717&logo=github&logoColor=whitee'></a> &nbsp;
54
  <a href='https://foundationagents.github.io/CARE/'><img src='https://img.shields.io/badge/%F0%9F%92%BB_Project-CARE-blue'></a> &nbsp;
55
  </div>
 
56
  ## Dataset Description
57
 
58
  This dataset is a processed version of DROP (Discrete Reasoning Over Paragraphs) adapted for training language models with native retrieval-augmented reasoning capabilities in the CARE framework. It serves as the "easy" dataset (D_easy) in the curriculum learning strategy for reinforcement learning training.
 
53
  <a href='https://github.com/FoundationAgents/CARE'><img src='https://img.shields.io/badge/_Code-CARE-181717?color=121717&logo=github&logoColor=whitee'></a> &nbsp;
54
  <a href='https://foundationagents.github.io/CARE/'><img src='https://img.shields.io/badge/%F0%9F%92%BB_Project-CARE-blue'></a> &nbsp;
55
  </div>
56
+
57
  ## Dataset Description
58
 
59
  This dataset is a processed version of DROP (Discrete Reasoning Over Paragraphs) adapted for training language models with native retrieval-augmented reasoning capabilities in the CARE framework. It serves as the "easy" dataset (D_easy) in the curriculum learning strategy for reinforcement learning training.