Yuanxh commited on
Commit
5f689ab
Β·
1 Parent(s): 80df33f

update official version

Browse files
Files changed (1) hide show
  1. constants.py +1 -1
constants.py CHANGED
@@ -32,7 +32,7 @@ XLSX_DIR = "./file//results.xlsx"
32
 
33
  LEADERBOARD_INTRODUCTION = """# πŸ† S-Eval Leaderboard
34
  ## πŸ”” Updates
35
- πŸ“£ [2025/03/30]: πŸŽ‰ Our paper has been accepted by ISSTA 2025. To meet evaluation needs under different budgets, we partition the benchmark into four scales: [Small](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/small) (1,000 Base and 10,000 Attack in each language), [Medium](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/medium) (3,000 Base and 30,000 Attack in each language), [Large](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/large) (5,000 Base and 50,000 Attack in each language) and [Full](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/full) (10,000 Base and 100,000 Attack in each language), comprehensively considering the balance and harmfulness of data.
36
 
37
  πŸ“£ [2024/10/25]: We release all 20,000 base risk prompts and 200,000 corresponding attack prompts ([Version-0.1.2](https://github.com/IS2Lab/S-Eval)). We also update [πŸ† LeaderBoard](https://huggingface.co/spaces/IS2Lab/S-Eval) with new evaluation results including GPT-4 and other models.
38
  πŸŽ‰ S-Eval has achieved about **7,000** total views and about **2,000** total downloads across multiple platforms. πŸŽ‰
 
32
 
33
  LEADERBOARD_INTRODUCTION = """# πŸ† S-Eval Leaderboard
34
  ## πŸ”” Updates
35
+ πŸ“£ [2025/03/30]: πŸŽ‰ Our [paper](https://dl.acm.org/doi/abs/10.1145/3728971) has been accepted by ISSTA 2025. To meet evaluation needs under different budgets, we partition the benchmark into four scales: [Small](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/small) (1,000 Base and 10,000 Attack in each language), [Medium](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/medium) (3,000 Base and 30,000 Attack in each language), [Large](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/large) (5,000 Base and 50,000 Attack in each language) and [Full](https://github.com/IS2Lab/S-Eval/tree/main/s_eval/full) (10,000 Base and 100,000 Attack in each language), comprehensively considering the balance and harmfulness of data.
36
 
37
  πŸ“£ [2024/10/25]: We release all 20,000 base risk prompts and 200,000 corresponding attack prompts ([Version-0.1.2](https://github.com/IS2Lab/S-Eval)). We also update [πŸ† LeaderBoard](https://huggingface.co/spaces/IS2Lab/S-Eval) with new evaluation results including GPT-4 and other models.
38
  πŸŽ‰ S-Eval has achieved about **7,000** total views and about **2,000** total downloads across multiple platforms. πŸŽ‰