FCE - Grammatical error detection
Detect errors in English text. [ver. 1.0.0]
This is a long list of all submissions, if you want to see only the best, click leaderboard.
| # | submitter | when | ver. | description | dev-0 Mean F0.5 | test-A Mean F0.5 | |
|---|---|---|---|---|---|---|---|
| 12 | s478846 | 2024-06-10 09:19 | 1.0.0 | mistral 7b oddballness again oddballness | 39.39 | 43.27 | |
| 17 | s478846 | 2024-01-24 12:05 | 1.0.0 | Roberta Large Oddballness (threshold 0.84) oddballness roberta | 42.23 | 40.78 | |
| 22 | s478846 | 2024-01-24 11:04 | 1.0.0 | Roberta Large Probability threshold 0.014 probability roberta | 37.08 | 34.51 | |
| 18 | s478846 | 2024-01-24 10:26 | 1.0.0 | Roberta Base oddballness threshold 0.91 oddballness roberta-base | 40.34 | 40.09 | |
| 8 | s478846 | 2024-01-24 10:17 | 1.0.0 | GPT2-small oddballness threshold 0.84 gpt2 oddballness | 46.96 | 45.52 | |
| 21 | s478846 | 2024-01-24 10:05 | 1.0.0 | Roberta Base Probability threshold 0.005 probability roberta-base | 37.91 | 36.95 | |
| 19 | s478846 | 2024-01-24 09:57 | 1.0.0 | GPT-2 small probabilities threshold 0.0002 gpt2 probability | 39.41 | 39.16 | |
| 3 | s478846 | 2023-12-15 11:11 | 1.0.0 | Max value GPT2-XL, Roberta-Large, Temperature 0.75, Threshold 0.986, Basic Tokenizer gpt2-xlarge oddballness roberta | 46.48 | 46.79 | |
| 16 | s478846 | 2023-11-10 13:55 | 1.0.0 | Llama 7b 16fp Probability (threshold 0.0004) probability | 39.39 | 40.81 | |
| 15 | s478846 | 2023-11-10 13:48 | 1.0.0 | Mistral 7b Probability (threshold 0.0003) probability | 40.78 | 41.19 | |
| 14 | s478846 | 2023-11-10 13:37 | 1.0.0 | Llama 7b 16fp Oddballness (threshold 0.84) oddballness | 42.25 | 42.93 | |
| 11 | s478846 | 2023-11-10 13:33 | 1.0.0 | Mistral 7b Oddballness (threshold 0.89) oddballness | 42.53 | 43.27 | |
| 20 | s478846 | 2023-11-09 09:56 | 1.0.0 | Yi-6B Probability (threshold 0.0005) probability | 38.58 | 38.79 | |
| 7 | s478846 | 2023-11-06 15:25 | 1.0.0 | Max value Yi-6B, RobertaLarge oddballness (threshold 0.91) oddballness roberta | 46.83 | 45.75 | |
| 5 | s478846 | 2023-11-06 14:47 | 1.0.0 | Yi-6B Oddballness (threshold 0.85) oddballness | 45.53 | 45.89 | |
| 13 | s478846 | 2023-11-06 12:14 | 1.0.0 | Probability min value from Roberta-Large, GPT2-XL (threshold 0.0001) gpt2-xlarge probability roberta | 44.93 | 43.15 | |
| 4 | s478846 | 2023-11-06 11:16 | 1.0.0 | GPT2-XL Oddballness (Threshold 0.85) gpt2-xlarge oddballness | 46.55 | 46.64 | |
| 10 | s478846 | 2023-10-30 16:00 | 1.0.0 | GPT2-XL Probabilities (Threshold 0.0001) probabilities gpt2-xlarge | 45.75 | 44.31 | |
| 23 | s478846 | 2023-10-30 15:51 | 1.0.0 | Roberta-Large probabilities (threshold 0.02) probabilities roberta | 36.76 | 33.80 | |
| 24 | s478846 | 2023-10-30 14:38 | 1.0.0 | All tokens as incorrect | 14.33 | 16.20 | |
| 2 | s478846 | 2023-10-30 14:30 | 1.0.0 | Max value GPT2-XL, Roberta-Large, Threshold 0.89, Basic Tokenizer gpt2-xlarge roberta | 46.51 | 47.05 | |
| 9 | s478846 | 2023-10-30 13:32 | 1.0.0 | Max value GPT-2, Roberta-Base Threshold 0.89 gpt2 roberta-base | 43.40 | 44.42 | |
| 6 | s478846 | 2023-10-30 10:54 | 1.0.0 | GPT-2 only threshold 0.89 gpt2 | 47.19 | 45.75 | |
| 1 | s478846 | 2023-10-28 19:36 | 1.0.0 | Max value from GPT2-XL, Roberta-Large gpt2-xlarge roberta | 47.09 | 47.30 |