Skip to content

GitLab

Switch branch/tag
Find file Normal viewHistoryPermalink
config_cn.json 2.07 KB
Newer Older
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
1
2
3
4 
{
  "language": "cn",
  "category": {
    "brainstorming": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
5 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
6
7
8
9
10
11
12
13
14
15
16 
        "language organization",
        "relevance",
        "creativity",
        "practicality",
        "correctness"
      ],
      "Metrics": [
        "Distinct"
      ]
    },
    "chat": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
17 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
18
19
20
21
22
23
24
25
26
27
28 
        "language organization",
        "relevance",
        "naturalness",
        "engagingness",
        "reasonableness"
      ],
      "Metrics": [
        "Distinct"
      ]
    },
    "classification": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
29 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
30
31
32
33
34
35
36
37
38
39
40 
        "language organization",
        "relevance",
        "correctness"
      ],
      "Metrics": [
        "Precision",
        "Recall",
        "F1 score"
      ]
    },
    "closed_qa": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
41 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
42
43
44
45
46
47
48
49
50
51
52 
        "language organization",
        "relevance",
        "correctness"
      ],
      "Metrics": [
        "BLEU",
        "ROUGE",
        "BERTScore"
      ]
    },
    "extraction": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
53 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
54
55
56
57
58
59
60
61
62
63
64 
        "language organization",
        "relevance",
        "correctness"
      ],
      "Metrics": [
        "Precision",
        "Recall",
        "F1 score"
      ]
    },
    "generation": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
65 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
66
67
68
69
70
71
72
73
74
75
76 
        "language organization",
        "relevance",
        "diversity"
      ],
      "Metrics": [
        "BLEU",
        "ROUGE",
        "BERTScore"
      ]
    },
    "open_qa": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
77 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
78
79
80
81
82
83
84
85
86 
        "language organization",
        "relevance",
        "correctness"
      ],
      "Metrics": [
        "Distinct"
      ]
    },
    "rewriting": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
87 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
88
89
90
91
92
93
94
95
96
97
98 
        "language organization",
        "relevance",
        "correctness"
      ],
      "Metrics": [
        "BLEU",
        "ROUGE",
        "BERTScore"
      ]
    },
    "roleplay": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
99 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
100
101
102
103
104
105
106
107
108
109 
        "language organization",
        "relevance",
        "fidelity",
        "creativity"
      ],
      "Metrics": [
        "Distinct"
      ]
    },
    "summarization": {
Yuanchen's avatar
[evaluation] improvement on evaluation (#3862)  
Yuanchen committed
110 
      "GPT": [
Yuanchen's avatar
[evaluation] add automatic evaluation pipeline (#3821)  
Yuanchen committed
111
112
113
114
115
116
117
118
119
120
121
122
123 
        "language organization",
        "relevance",
        "correctness",
        "conciseness"
      ],
      "Metrics": [
        "BLEU",
        "ROUGE",
        "BERTScore"
      ]
    }
  }
}