Datasets:
KaifengGGG
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -155,29 +155,6 @@ configs:
|
|
155 |
|
156 |
The WenYanWen\_English\_Parallel dataset is a multilingual parallel corpus in Classical Chinese (Wenyanwen), modern Chinese, and English. The Classical Chinese and modern Chinese parts are sourced from the NiuTrans/Classical-Modern dataset, while the corresponding English translations are generated using Gemini Pro.
|
157 |
|
158 |
-
## **Supported Tasks and Leaderboard**
|
159 |
-
|
160 |
-
This dataset can be used for various multilingual and translation tasks, including but not limited to:
|
161 |
-
|
162 |
-
1. Neural Machine Translation (Classical Chinese to Modern Chinese)
|
163 |
-
2. Neural Machine Translation (Modern Chinese to English)
|
164 |
-
3. Neural Machine Translation (Classical Chinese to English)
|
165 |
-
4. Multilingual Text-to-Text Transfer
|
166 |
-
|
167 |
-
There is currently no official leaderboard for this dataset.
|
168 |
-
|
169 |
-
## **License**
|
170 |
-
|
171 |
-
Please refer to the license of the [NiuTrans/Classical-Modern](https://github.com/NiuTrans/Classical-Modern) dataset and the terms of use of Gemini Pro for more information regarding the dataset license.
|
172 |
-
|
173 |
-
## **Citation Information**
|
174 |
-
|
175 |
-
If you use this dataset in your research, please cite the original sources:
|
176 |
-
|
177 |
-
1. [NiuTrans/Classical-Modern](https://github.com/NiuTrans/Classical-Modern)
|
178 |
-
2. [Gemini Pro](https://arxiv.org/abs/2403.05530)
|
179 |
-
|
180 |
-
|
181 |
## **Data Fields**
|
182 |
- `info`: A string representing the title or source information of the text.
|
183 |
- `classical`: Classical Chinese (Wenyanwen) text corresponding to the modern text.
|
@@ -186,7 +163,7 @@ If you use this dataset in your research, please cite the original sources:
|
|
186 |
- `text`: instruction/answer pair in string format
|
187 |
- `messages`: instruction/answer pair in conversation format:
|
188 |
- `content`: String representing the content of a message.
|
189 |
-
- `role`: String representing the role associated with the message (e.g.,
|
190 |
|
191 |
Here is an example for a dataset entry:
|
192 |
|
@@ -203,69 +180,66 @@ Here is an example for a dataset entry:
|
|
203 |
|
204 |
The dataset have four subsets, namely `default`, `instruct`, `instruct-augment`, `instruct-large`
|
205 |
### **Default**
|
206 |
-
|
207 |
-
|
208 |
-
|
209 |
-
- `classical`: string
|
210 |
-
- `english`: string
|
211 |
-
|
212 |
-
**Splits:**
|
213 |
-
- Train: 972,467 examples
|
214 |
|
215 |
-
**Size:**
|
216 |
-
- Dataset size: 366.9 MB
|
217 |
-
- Download size: 256.4 MB
|
218 |
|
219 |
-
|
220 |
-
|
221 |
-
|
222 |
-
- `modern`: string
|
223 |
-
- `classical`: string
|
224 |
-
- `english`: string
|
225 |
-
- `text`: string
|
226 |
-
- `messages`: list of {`content`: string, `role`: string}
|
227 |
|
228 |
-
|
229 |
-
|
230 |
-
|
|
|
231 |
|
232 |
-
|
233 |
-
|
234 |
-
|
|
|
235 |
|
236 |
### **Instruct-Augmented**
|
237 |
-
|
238 |
-
|
239 |
-
|
240 |
-
|
241 |
-
|
242 |
-
|
243 |
-
|
244 |
-
|
245 |
-
**Splits:**
|
246 |
-
- Train: 9,000 examples
|
247 |
-
- Test: 1,000 examples
|
248 |
-
|
249 |
-
**Size:**
|
250 |
-
- Dataset size: 12.4 MB
|
251 |
-
- Download size: 7.56 MB
|
252 |
|
253 |
### **Instruct-Large**
|
254 |
-
|
255 |
-
|
256 |
-
|
257 |
-
- `classical`: string
|
258 |
-
- `english`: string
|
259 |
-
- `text`: string
|
260 |
-
- `messages`: list of {`content`: string, `role`: string}
|
261 |
|
262 |
-
|
263 |
-
|
264 |
-
|
|
|
265 |
|
266 |
-
|
267 |
-
|
268 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
269 |
|
270 |
## **Potential Bias**
|
271 |
|
@@ -273,4 +247,6 @@ Since the English translations are generated using Gemini Pro, there might be in
|
|
273 |
|
274 |
## **Potential Social Impact**
|
275 |
|
276 |
-
This dataset can be used for various multilingual and translation tasks, which can have a positive impact on facilitating cross-cultural communication and understanding. However, it is important to be aware of the potential biases in the dataset and to use the dataset responsibly. Additionally, as with any dataset, it is important to consider the ethical implications of using this dataset, including issues related to data privacy, consent, and representation.
|
|
|
|
|
|
155 |
|
156 |
The WenYanWen\_English\_Parallel dataset is a multilingual parallel corpus in Classical Chinese (Wenyanwen), modern Chinese, and English. The Classical Chinese and modern Chinese parts are sourced from the NiuTrans/Classical-Modern dataset, while the corresponding English translations are generated using Gemini Pro.
|
157 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
158 |
## **Data Fields**
|
159 |
- `info`: A string representing the title or source information of the text.
|
160 |
- `classical`: Classical Chinese (Wenyanwen) text corresponding to the modern text.
|
|
|
163 |
- `text`: instruction/answer pair in string format
|
164 |
- `messages`: instruction/answer pair in conversation format:
|
165 |
- `content`: String representing the content of a message.
|
166 |
+
- `role`: String representing the role associated with the message (e.g., system, assistent, user).
|
167 |
|
168 |
Here is an example for a dataset entry:
|
169 |
|
|
|
180 |
|
181 |
The dataset have four subsets, namely `default`, `instruct`, `instruct-augment`, `instruct-large`
|
182 |
### **Default**
|
183 |
+
| `info` | `modern` | `classical` | `english` |
|
184 |
+
|----------|----------|-------------|-----------|
|
185 |
+
| string | string | string | string |
|
|
|
|
|
|
|
|
|
|
|
186 |
|
|
|
|
|
|
|
187 |
|
188 |
+
| Split | Examples |
|
189 |
+
|-------|-----------|
|
190 |
+
| Train | 972,467 |
|
|
|
|
|
|
|
|
|
|
|
191 |
|
192 |
+
### **Instruct**
|
193 |
+
| `info` | `modern` | `classical` | `english` | `text` | `messages` |
|
194 |
+
|----------|----------|-------------|-----------|--------|------------------------|
|
195 |
+
| string | string | string | string | string | list of {`content`: string, `role`: string}|
|
196 |
|
197 |
+
| Split | Examples |
|
198 |
+
|-------|-----------|
|
199 |
+
| Train | 9,000 |
|
200 |
+
| Test | 1,000 |
|
201 |
|
202 |
### **Instruct-Augmented**
|
203 |
+
| `info` | `modern` | `classical` | `english` | `text` | `messages` |
|
204 |
+
|----------|----------|-------------|-----------|--------|------------------------|
|
205 |
+
| string | string | string | string | string | list of {`content`: string, `role`: string}|
|
206 |
+
|
207 |
+
| Split | Examples |
|
208 |
+
|-------|-----------|
|
209 |
+
| Train | 9,000 |
|
210 |
+
| Test | 1,000 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
211 |
|
212 |
### **Instruct-Large**
|
213 |
+
| `info` | `modern` | `classical` | `english` | `text` | `messages` |
|
214 |
+
|----------|----------|-------------|-----------|--------|------------------------|
|
215 |
+
| string | string | string | string | string | list of {`content`: string, `role`: string}|
|
|
|
|
|
|
|
|
|
216 |
|
217 |
+
| Split | Examples |
|
218 |
+
|-------|-----------|
|
219 |
+
| Train | 875,214 |
|
220 |
+
| Test | 97,246 |
|
221 |
|
222 |
+
## **Supported Tasks and Leaderboard**
|
223 |
+
|
224 |
+
This dataset can be used for various multilingual and translation tasks, including but not limited to:
|
225 |
+
|
226 |
+
1. Neural Machine Translation (Classical Chinese to Modern Chinese)
|
227 |
+
2. Neural Machine Translation (Modern Chinese to English)
|
228 |
+
3. Neural Machine Translation (Classical Chinese to English)
|
229 |
+
4. Multilingual Text-to-Text Transfer
|
230 |
+
|
231 |
+
There is currently no official leaderboard for this dataset.
|
232 |
+
|
233 |
+
## **License**
|
234 |
+
|
235 |
+
Please refer to the license of the [NiuTrans/Classical-Modern](https://github.com/NiuTrans/Classical-Modern) dataset and the terms of use of Gemini Pro for more information regarding the dataset license.
|
236 |
+
|
237 |
+
## **Citation Information**
|
238 |
+
|
239 |
+
If you use this dataset in your research, please cite the original sources:
|
240 |
+
|
241 |
+
1. [NiuTrans/Classical-Modern](https://github.com/NiuTrans/Classical-Modern)
|
242 |
+
2. [Gemini Pro](https://arxiv.org/abs/2403.05530)
|
243 |
|
244 |
## **Potential Bias**
|
245 |
|
|
|
247 |
|
248 |
## **Potential Social Impact**
|
249 |
|
250 |
+
This dataset can be used for various multilingual and translation tasks, which can have a positive impact on facilitating cross-cultural communication and understanding. However, it is important to be aware of the potential biases in the dataset and to use the dataset responsibly. Additionally, as with any dataset, it is important to consider the ethical implications of using this dataset, including issues related to data privacy, consent, and representation.
|
251 |
+
|
252 |
+
|