Ihor commited on
Commit
18e0e43
·
verified ·
1 Parent(s): 913341c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -9
README.md CHANGED
@@ -212,7 +212,7 @@ for summary in summaries:
212
 
213
  ![Model Performance](gliner_multitask_performance.png)
214
 
215
- Our multitask model demonstrates comparable performance on different zero-shot benchmarks to dedicated models to NER task:
216
 
217
  | Model | Dataset | Precision | Recall | F1 Score | F1 Score (Decimal) |
218
  |------------------------------------|--------------------|-----------|--------|----------|--------------------|
@@ -222,32 +222,32 @@ Our multitask model demonstrates comparable performance on different zero-shot b
222
  | | CrossNER_politics | 77.28% | 68.69% | 72.73% | 0.7273 |
223
  | | CrossNER_science | 70.08% | 63.12% | 66.42% | 0.6642 |
224
  | | mit-movie | 63.00% | 48.88% | 55.05% | 0.5505 |
225
- | | mit-restaurant | 51.30% | 34.35% | 41.15% | 0.4115 |
226
- | | **Average** | | | | **0.6147** |
227
  | knowledgator/gliner-multitask-v0.5 | CrossNER_AI | 51.00% | 51.11% | 51.05% | 0.5105 |
228
  | | CrossNER_literature | 72.65% | 65.62% | 68.96% | 0.6896 |
229
  | | CrossNER_music | 74.91% | 73.70% | 74.30% | 0.7430 |
230
  | | CrossNER_politics | 78.84% | 77.71% | 78.27% | 0.7827 |
231
  | | CrossNER_science | 69.20% | 65.48% | 67.29% | 0.6729 |
232
  | | mit-movie | 61.29% | 52.59% | 56.60% | 0.5660 |
233
- | | mit-restaurant | 50.65% | 36.95% | 42.73% | 0.4273 |
234
- | | **Average** | | | | **0.6267** |
235
  | urchade/gliner_large-v2.1 | CrossNER_AI | 54.98% | 52.00% | 53.45% | 0.5345 |
236
  | | CrossNER_literature| 59.33% | 56.47% | 57.87% | 0.5787 |
237
  | | CrossNER_music | 67.39% | 66.77% | 67.08% | 0.6708 |
238
  | | CrossNER_politics | 66.07% | 63.76% | 64.90% | 0.6490 |
239
  | | CrossNER_science | 61.45% | 62.56% | 62.00% | 0.6200 |
240
  | | mit-movie | 55.94% | 47.36% | 51.29% | 0.5129 |
241
- | | mit-restaurant | 52.89% | 40.63% | 45.96% | 0.4596 |
242
- | | **Average** | | | | **0.5751** |
243
  | EmergentMethods/gliner_large_news-v2.1| CrossNER_AI | 59.60% | 54.55% | 56.96% | 0.5696 |
244
  | | CrossNER_literature| 65.41% | 56.16% | 60.44% | 0.6044 |
245
  | | CrossNER_music | 67.47% | 63.08% | 65.20% | 0.6520 |
246
  | | CrossNER_politics | 66.05% | 60.07% | 62.92% | 0.6292 |
247
  | | CrossNER_science | 68.44% | 63.57% | 65.92% | 0.6592 |
248
  | | mit-movie | 65.85% | 49.59% | 56.57% | 0.5657 |
249
- | | mit-restaurant | 51.54% | 35.11% | 41.77% | 0.4177 |
250
- | | **Average** | | | | **0.5854** |
251
 
252
 
253
  ### Join Our Discord
 
212
 
213
  ![Model Performance](gliner_multitask_performance.png)
214
 
215
+ Our multitask model demonstrates comparable performance on different zero-shot benchmarks to dedicated models to NER task (all labels were lowecased in this testing):
216
 
217
  | Model | Dataset | Precision | Recall | F1 Score | F1 Score (Decimal) |
218
  |------------------------------------|--------------------|-----------|--------|----------|--------------------|
 
222
  | | CrossNER_politics | 77.28% | 68.69% | 72.73% | 0.7273 |
223
  | | CrossNER_science | 70.08% | 63.12% | 66.42% | 0.6642 |
224
  | | mit-movie | 63.00% | 48.88% | 55.05% | 0.5505 |
225
+ | | mit-restaurant | 54.81% | 37.62% | 44.62% | 0.4462 |
226
+ | | **Average** | | | | **0.6196** |
227
  | knowledgator/gliner-multitask-v0.5 | CrossNER_AI | 51.00% | 51.11% | 51.05% | 0.5105 |
228
  | | CrossNER_literature | 72.65% | 65.62% | 68.96% | 0.6896 |
229
  | | CrossNER_music | 74.91% | 73.70% | 74.30% | 0.7430 |
230
  | | CrossNER_politics | 78.84% | 77.71% | 78.27% | 0.7827 |
231
  | | CrossNER_science | 69.20% | 65.48% | 67.29% | 0.6729 |
232
  | | mit-movie | 61.29% | 52.59% | 56.60% | 0.5660 |
233
+ | | mit-restaurant | 50.65% | 38.13% | 43.51% | 0.4351 |
234
+ | | **Average** | | | | **0.6276** |
235
  | urchade/gliner_large-v2.1 | CrossNER_AI | 54.98% | 52.00% | 53.45% | 0.5345 |
236
  | | CrossNER_literature| 59.33% | 56.47% | 57.87% | 0.5787 |
237
  | | CrossNER_music | 67.39% | 66.77% | 67.08% | 0.6708 |
238
  | | CrossNER_politics | 66.07% | 63.76% | 64.90% | 0.6490 |
239
  | | CrossNER_science | 61.45% | 62.56% | 62.00% | 0.6200 |
240
  | | mit-movie | 55.94% | 47.36% | 51.29% | 0.5129 |
241
+ | | mit-restaurant | 53.34% | 40.83% | 46.25% | 0.4625 |
242
+ | | **Average** | | | | **0.5754** |
243
  | EmergentMethods/gliner_large_news-v2.1| CrossNER_AI | 59.60% | 54.55% | 56.96% | 0.5696 |
244
  | | CrossNER_literature| 65.41% | 56.16% | 60.44% | 0.6044 |
245
  | | CrossNER_music | 67.47% | 63.08% | 65.20% | 0.6520 |
246
  | | CrossNER_politics | 66.05% | 60.07% | 62.92% | 0.6292 |
247
  | | CrossNER_science | 68.44% | 63.57% | 65.92% | 0.6592 |
248
  | | mit-movie | 65.85% | 49.59% | 56.57% | 0.5657 |
249
+ | | mit-restaurant | 54.71% | 35.94% | 43.38% | 0.4338 |
250
+ | | **Average** | | | | **0.5876** |
251
 
252
 
253
  ### Join Our Discord