QShane commited on
Commit
550d5da
·
verified ·
1 Parent(s): f6dc2dc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -2
README.md CHANGED
@@ -2,8 +2,6 @@
2
 
3
  # CL²GEC:A Multi-Discipline Benchmark for Continual Learning in Chinese Literature Grammatical Error Correction
4
 
5
- ## Dataset Summary
6
-
7
  **CL²GEC** is a benchmark for **Chinese grammatical error correction (GEC)** in **scholarly writing** with a **continual-learning** protocol. The corpus covers **10 first-level disciplines** (Law, Management, Education, Economics, Natural Sciences, History, Agricultural Sciences, Literature, Arts, Philosophy). Each sample contains an errorful sentence (`source`) and one or more corrected references (`references`). Standard **train / validation / test** splits are provided and may be used **per-discipline** to study sequential/continual learning behavior such as forgetting and transfer.
8
 
9
  ---
 
2
 
3
  # CL²GEC:A Multi-Discipline Benchmark for Continual Learning in Chinese Literature Grammatical Error Correction
4
 
 
 
5
  **CL²GEC** is a benchmark for **Chinese grammatical error correction (GEC)** in **scholarly writing** with a **continual-learning** protocol. The corpus covers **10 first-level disciplines** (Law, Management, Education, Economics, Natural Sciences, History, Agricultural Sciences, Literature, Arts, Philosophy). Each sample contains an errorful sentence (`source`) and one or more corrected references (`references`). Standard **train / validation / test** splits are provided and may be used **per-discipline** to study sequential/continual learning behavior such as forgetting and transfer.
6
 
7
  ---