PEFT
qlora
sft
trl
qwen3
tmf921
intent-based-networking
network-slicing
rtx-6000-ada
ml-intern
nraptisss commited on
Commit
de75d5b
·
verified ·
1 Parent(s): e0c5f96

Upload folder using huggingface_hub

Browse files
results/baselines/qwen3_8b_zero_shot_normalized_200.json ADDED
@@ -0,0 +1,2378 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "test_adversarial": {
3
+ "num_examples": 33,
4
+ "parse_json": 0.0,
5
+ "gold_parse_json": 1.0,
6
+ "exact_match": 0.0,
7
+ "field_precision": 0.0,
8
+ "field_recall": 0.0,
9
+ "field_f1": 0.0,
10
+ "field_tp": 0.0,
11
+ "field_fp": 0.0,
12
+ "field_fn": 0.0,
13
+ "slice_sst_pass": 0.0,
14
+ "kpi_text_presence_pass": 0.0,
15
+ "adversarial_status_pass": 0.0,
16
+ "norm_parse_json": 0.0,
17
+ "norm_gold_parse_json": 1.0,
18
+ "norm_exact_match": 0.0,
19
+ "norm_field_precision": 0.0,
20
+ "norm_field_recall": 0.0,
21
+ "norm_field_f1": 0.0,
22
+ "norm_key_precision": 0.0,
23
+ "norm_key_recall": 0.0,
24
+ "norm_key_f1": 0.0,
25
+ "by_target_layer": {
26
+ "adversarial_ambiguous": {
27
+ "num_examples": 17,
28
+ "parse_json": 0.0,
29
+ "gold_parse_json": 1.0,
30
+ "exact_match": 0.0,
31
+ "field_precision": 0.0,
32
+ "field_recall": 0.0,
33
+ "field_f1": 0.0,
34
+ "field_tp": 0.0,
35
+ "field_fp": 0.0,
36
+ "field_fn": 0.0,
37
+ "slice_sst_pass": 0.0,
38
+ "kpi_text_presence_pass": 0.0,
39
+ "adversarial_status_pass": 0.0,
40
+ "norm_parse_json": 0.0,
41
+ "norm_gold_parse_json": 1.0,
42
+ "norm_exact_match": 0.0,
43
+ "norm_field_precision": 0.0,
44
+ "norm_field_recall": 0.0,
45
+ "norm_field_f1": 0.0,
46
+ "norm_key_precision": 0.0,
47
+ "norm_key_recall": 0.0,
48
+ "norm_key_f1": 0.0
49
+ },
50
+ "adversarial_contradictory": {
51
+ "num_examples": 9,
52
+ "parse_json": 0.0,
53
+ "gold_parse_json": 1.0,
54
+ "exact_match": 0.0,
55
+ "field_precision": 0.0,
56
+ "field_recall": 0.0,
57
+ "field_f1": 0.0,
58
+ "field_tp": 0.0,
59
+ "field_fp": 0.0,
60
+ "field_fn": 0.0,
61
+ "slice_sst_pass": 0.0,
62
+ "kpi_text_presence_pass": 0.0,
63
+ "adversarial_status_pass": 0.0,
64
+ "norm_parse_json": 0.0,
65
+ "norm_gold_parse_json": 1.0,
66
+ "norm_exact_match": 0.0,
67
+ "norm_field_precision": 0.0,
68
+ "norm_field_recall": 0.0,
69
+ "norm_field_f1": 0.0,
70
+ "norm_key_precision": 0.0,
71
+ "norm_key_recall": 0.0,
72
+ "norm_key_f1": 0.0
73
+ },
74
+ "adversarial_out_of_scope": {
75
+ "num_examples": 7,
76
+ "parse_json": 0.0,
77
+ "gold_parse_json": 1.0,
78
+ "exact_match": 0.0,
79
+ "field_precision": 0.0,
80
+ "field_recall": 0.0,
81
+ "field_f1": 0.0,
82
+ "field_tp": 0.0,
83
+ "field_fp": 0.0,
84
+ "field_fn": 0.0,
85
+ "slice_sst_pass": 0.0,
86
+ "kpi_text_presence_pass": 0.0,
87
+ "adversarial_status_pass": 0.0,
88
+ "norm_parse_json": 0.0,
89
+ "norm_gold_parse_json": 1.0,
90
+ "norm_exact_match": 0.0,
91
+ "norm_field_precision": 0.0,
92
+ "norm_field_recall": 0.0,
93
+ "norm_field_f1": 0.0,
94
+ "norm_key_precision": 0.0,
95
+ "norm_key_recall": 0.0,
96
+ "norm_key_f1": 0.0
97
+ }
98
+ },
99
+ "by_slice_type": {
100
+ "N/A": {
101
+ "num_examples": 33,
102
+ "parse_json": 0.0,
103
+ "gold_parse_json": 1.0,
104
+ "exact_match": 0.0,
105
+ "field_precision": 0.0,
106
+ "field_recall": 0.0,
107
+ "field_f1": 0.0,
108
+ "field_tp": 0.0,
109
+ "field_fp": 0.0,
110
+ "field_fn": 0.0,
111
+ "slice_sst_pass": 0.0,
112
+ "kpi_text_presence_pass": 0.0,
113
+ "adversarial_status_pass": 0.0,
114
+ "norm_parse_json": 0.0,
115
+ "norm_gold_parse_json": 1.0,
116
+ "norm_exact_match": 0.0,
117
+ "norm_field_precision": 0.0,
118
+ "norm_field_recall": 0.0,
119
+ "norm_field_f1": 0.0,
120
+ "norm_key_precision": 0.0,
121
+ "norm_key_recall": 0.0,
122
+ "norm_key_f1": 0.0
123
+ }
124
+ },
125
+ "by_lifecycle_operation": {
126
+ "create": {
127
+ "num_examples": 33,
128
+ "parse_json": 0.0,
129
+ "gold_parse_json": 1.0,
130
+ "exact_match": 0.0,
131
+ "field_precision": 0.0,
132
+ "field_recall": 0.0,
133
+ "field_f1": 0.0,
134
+ "field_tp": 0.0,
135
+ "field_fp": 0.0,
136
+ "field_fn": 0.0,
137
+ "slice_sst_pass": 0.0,
138
+ "kpi_text_presence_pass": 0.0,
139
+ "adversarial_status_pass": 0.0,
140
+ "norm_parse_json": 0.0,
141
+ "norm_gold_parse_json": 1.0,
142
+ "norm_exact_match": 0.0,
143
+ "norm_field_precision": 0.0,
144
+ "norm_field_recall": 0.0,
145
+ "norm_field_f1": 0.0,
146
+ "norm_key_precision": 0.0,
147
+ "norm_key_recall": 0.0,
148
+ "norm_key_f1": 0.0
149
+ }
150
+ }
151
+ },
152
+ "test_in_distribution": {
153
+ "num_examples": 200,
154
+ "parse_json": 0.335,
155
+ "gold_parse_json": 1.0,
156
+ "exact_match": 0.0,
157
+ "field_precision": 0.0015219907407407406,
158
+ "field_recall": 0.0008423913043478261,
159
+ "field_f1": 0.0010584844594233136,
160
+ "field_tp": 0.03,
161
+ "field_fp": 5.855,
162
+ "field_fn": 10.515,
163
+ "slice_sst_pass": 0.285,
164
+ "kpi_text_presence_pass": 0.275,
165
+ "adversarial_status_pass": 0.335,
166
+ "norm_parse_json": 0.335,
167
+ "norm_gold_parse_json": 1.0,
168
+ "norm_exact_match": 0.0,
169
+ "norm_field_precision": 0.0012789351851851853,
170
+ "norm_field_recall": 0.0007763157894736841,
171
+ "norm_field_f1": 0.0009395604395604396,
172
+ "norm_key_precision": 0.026960091567838806,
173
+ "norm_key_recall": 0.012895510835913313,
174
+ "norm_key_f1": 0.016929998260302812,
175
+ "by_target_layer": {
176
+ "a1_policy": {
177
+ "num_examples": 27,
178
+ "parse_json": 0.14814814814814814,
179
+ "gold_parse_json": 1.0,
180
+ "exact_match": 0.0,
181
+ "field_precision": 0.0,
182
+ "field_recall": 0.0,
183
+ "field_f1": 0.0,
184
+ "field_tp": 0.0,
185
+ "field_fp": 1.5555555555555556,
186
+ "field_fn": 2.814814814814815,
187
+ "slice_sst_pass": 0.1111111111111111,
188
+ "kpi_text_presence_pass": 0.07407407407407407,
189
+ "adversarial_status_pass": 0.14814814814814814,
190
+ "norm_parse_json": 0.14814814814814814,
191
+ "norm_gold_parse_json": 1.0,
192
+ "norm_exact_match": 0.0,
193
+ "norm_field_precision": 0.0,
194
+ "norm_field_recall": 0.0,
195
+ "norm_field_f1": 0.0,
196
+ "norm_key_precision": 0.0,
197
+ "norm_key_recall": 0.0,
198
+ "norm_key_f1": 0.0
199
+ },
200
+ "camara": {
201
+ "num_examples": 36,
202
+ "parse_json": 0.7222222222222222,
203
+ "gold_parse_json": 1.0,
204
+ "exact_match": 0.0,
205
+ "field_precision": 0.0,
206
+ "field_recall": 0.0,
207
+ "field_f1": 0.0,
208
+ "field_tp": 0.0,
209
+ "field_fp": 5.777777777777778,
210
+ "field_fn": 11.61111111111111,
211
+ "slice_sst_pass": 0.5833333333333334,
212
+ "kpi_text_presence_pass": 0.7222222222222222,
213
+ "adversarial_status_pass": 0.7222222222222222,
214
+ "norm_parse_json": 0.7222222222222222,
215
+ "norm_gold_parse_json": 1.0,
216
+ "norm_exact_match": 0.0,
217
+ "norm_field_precision": 0.0,
218
+ "norm_field_recall": 0.0,
219
+ "norm_field_f1": 0.0,
220
+ "norm_field_tp": 0.0,
221
+ "norm_field_fp": 8.0,
222
+ "norm_field_fn": 14.076923076923077,
223
+ "norm_key_precision": 0.0,
224
+ "norm_key_recall": 0.0,
225
+ "norm_key_f1": 0.0,
226
+ "norm_key_tp": 0.0,
227
+ "norm_key_fp": 8.0,
228
+ "norm_key_fn": 14.076923076923077
229
+ },
230
+ "etsi_zsm": {
231
+ "num_examples": 20,
232
+ "parse_json": 0.25,
233
+ "gold_parse_json": 1.0,
234
+ "exact_match": 0.0,
235
+ "field_precision": 0.0,
236
+ "field_recall": 0.0,
237
+ "field_f1": 0.0,
238
+ "field_tp": 0.0,
239
+ "field_fp": 8.85,
240
+ "field_fn": 14.25,
241
+ "slice_sst_pass": 0.25,
242
+ "kpi_text_presence_pass": 0.25,
243
+ "adversarial_status_pass": 0.25,
244
+ "norm_parse_json": 0.25,
245
+ "norm_gold_parse_json": 1.0,
246
+ "norm_exact_match": 0.0,
247
+ "norm_field_precision": 0.0,
248
+ "norm_field_recall": 0.0,
249
+ "norm_field_f1": 0.0,
250
+ "norm_key_precision": 0.0,
251
+ "norm_key_recall": 0.0,
252
+ "norm_key_f1": 0.0
253
+ },
254
+ "intent_3gpp": {
255
+ "num_examples": 39,
256
+ "parse_json": 0.46153846153846156,
257
+ "gold_parse_json": 1.0,
258
+ "exact_match": 0.0,
259
+ "field_precision": 0.004599952516619183,
260
+ "field_recall": 0.003205128205128205,
261
+ "field_f1": 0.003773865714164222,
262
+ "field_tp": 0.1282051282051282,
263
+ "field_fp": 12.333333333333334,
264
+ "field_fn": 18.333333333333332,
265
+ "slice_sst_pass": 0.358974358974359,
266
+ "kpi_text_presence_pass": 0.38461538461538464,
267
+ "adversarial_status_pass": 0.46153846153846156,
268
+ "norm_parse_json": 0.46153846153846156,
269
+ "norm_gold_parse_json": 1.0,
270
+ "norm_exact_match": 0.0,
271
+ "norm_field_precision": 0.0033535137701804366,
272
+ "norm_field_recall": 0.002699055330634278,
273
+ "norm_field_f1": 0.002986756832910679,
274
+ "norm_key_precision": 0.03392949615986106,
275
+ "norm_key_recall": 0.026990553306342778,
276
+ "norm_key_f1": 0.03000440808822427
277
+ },
278
+ "o1_nrm": {
279
+ "num_examples": 19,
280
+ "parse_json": 0.21052631578947367,
281
+ "gold_parse_json": 1.0,
282
+ "exact_match": 0.0,
283
+ "field_precision": 0.006578947368421052,
284
+ "field_recall": 0.002288329519450801,
285
+ "field_f1": 0.003395585738539898,
286
+ "field_tp": 0.05263157894736842,
287
+ "field_fp": 2.0526315789473686,
288
+ "field_fn": 4.7894736842105265,
289
+ "slice_sst_pass": 0.21052631578947367,
290
+ "kpi_text_presence_pass": 0.05263157894736842,
291
+ "adversarial_status_pass": 0.21052631578947367,
292
+ "norm_parse_json": 0.21052631578947367,
293
+ "norm_gold_parse_json": 1.0,
294
+ "norm_exact_match": 0.0,
295
+ "norm_field_precision": 0.006578947368421052,
296
+ "norm_field_recall": 0.002631578947368421,
297
+ "norm_field_f1": 0.003759398496240602,
298
+ "norm_key_precision": 0.18890977443609022,
299
+ "norm_key_recall": 0.07105263157894737,
300
+ "norm_key_f1": 0.10317460317460317
301
+ },
302
+ "tmf921": {
303
+ "num_examples": 45,
304
+ "parse_json": 0.13333333333333333,
305
+ "gold_parse_json": 1.0,
306
+ "exact_match": 0.0,
307
+ "field_precision": 0.0,
308
+ "field_recall": 0.0,
309
+ "field_f1": 0.0,
310
+ "field_tp": 0.0,
311
+ "field_fp": 4.5777777777777775,
312
+ "field_fn": 10.533333333333333,
313
+ "slice_sst_pass": 0.13333333333333333,
314
+ "kpi_text_presence_pass": 0.13333333333333333,
315
+ "adversarial_status_pass": 0.13333333333333333,
316
+ "norm_parse_json": 0.13333333333333333,
317
+ "norm_gold_parse_json": 1.0,
318
+ "norm_exact_match": 0.0,
319
+ "norm_field_precision": 0.0,
320
+ "norm_field_recall": 0.0,
321
+ "norm_field_f1": 0.0,
322
+ "norm_key_precision": 0.010655161089943698,
323
+ "norm_key_recall": 0.00392156862745098,
324
+ "norm_key_f1": 0.005678006140052345
325
+ },
326
+ "tmf921_lifecycle_activate": {
327
+ "num_examples": 2,
328
+ "parse_json": 0.0,
329
+ "gold_parse_json": 1.0,
330
+ "exact_match": 0.0,
331
+ "field_precision": 0.0,
332
+ "field_recall": 0.0,
333
+ "field_f1": 0.0,
334
+ "field_tp": 0.0,
335
+ "field_fp": 0.0,
336
+ "field_fn": 0.0,
337
+ "slice_sst_pass": 0.0,
338
+ "kpi_text_presence_pass": 0.0,
339
+ "adversarial_status_pass": 0.0,
340
+ "norm_parse_json": 0.0,
341
+ "norm_gold_parse_json": 1.0,
342
+ "norm_exact_match": 0.0,
343
+ "norm_field_precision": 0.0,
344
+ "norm_field_recall": 0.0,
345
+ "norm_field_f1": 0.0,
346
+ "norm_key_precision": 0.0,
347
+ "norm_key_recall": 0.0,
348
+ "norm_key_f1": 0.0
349
+ },
350
+ "tmf921_lifecycle_report": {
351
+ "num_examples": 4,
352
+ "parse_json": 0.0,
353
+ "gold_parse_json": 1.0,
354
+ "exact_match": 0.0,
355
+ "field_precision": 0.0,
356
+ "field_recall": 0.0,
357
+ "field_f1": 0.0,
358
+ "field_tp": 0.0,
359
+ "field_fp": 0.0,
360
+ "field_fn": 0.0,
361
+ "slice_sst_pass": 0.0,
362
+ "kpi_text_presence_pass": 0.0,
363
+ "adversarial_status_pass": 0.0,
364
+ "norm_parse_json": 0.0,
365
+ "norm_gold_parse_json": 1.0,
366
+ "norm_exact_match": 0.0,
367
+ "norm_field_precision": 0.0,
368
+ "norm_field_recall": 0.0,
369
+ "norm_field_f1": 0.0,
370
+ "norm_key_precision": 0.0,
371
+ "norm_key_recall": 0.0,
372
+ "norm_key_f1": 0.0
373
+ },
374
+ "tmf921_lifecycle_resume": {
375
+ "num_examples": 2,
376
+ "parse_json": 0.0,
377
+ "gold_parse_json": 1.0,
378
+ "exact_match": 0.0,
379
+ "field_precision": 0.0,
380
+ "field_recall": 0.0,
381
+ "field_f1": 0.0,
382
+ "field_tp": 0.0,
383
+ "field_fp": 0.0,
384
+ "field_fn": 0.0,
385
+ "slice_sst_pass": 0.0,
386
+ "kpi_text_presence_pass": 0.0,
387
+ "adversarial_status_pass": 0.0,
388
+ "norm_parse_json": 0.0,
389
+ "norm_gold_parse_json": 1.0,
390
+ "norm_exact_match": 0.0,
391
+ "norm_field_precision": 0.0,
392
+ "norm_field_recall": 0.0,
393
+ "norm_field_f1": 0.0,
394
+ "norm_key_precision": 0.0,
395
+ "norm_key_recall": 0.0,
396
+ "norm_key_f1": 0.0
397
+ },
398
+ "tmf921_lifecycle_scale": {
399
+ "num_examples": 3,
400
+ "parse_json": 0.6666666666666666,
401
+ "gold_parse_json": 1.0,
402
+ "exact_match": 0.0,
403
+ "field_precision": 0.0,
404
+ "field_recall": 0.0,
405
+ "field_f1": 0.0,
406
+ "field_tp": 0.0,
407
+ "field_fp": 3.6666666666666665,
408
+ "field_fn": 10.0,
409
+ "slice_sst_pass": 0.6666666666666666,
410
+ "kpi_text_presence_pass": 0.0,
411
+ "adversarial_status_pass": 0.6666666666666666,
412
+ "norm_parse_json": 0.6666666666666666,
413
+ "norm_gold_parse_json": 1.0,
414
+ "norm_exact_match": 0.0,
415
+ "norm_field_precision": 0.0,
416
+ "norm_field_recall": 0.0,
417
+ "norm_field_f1": 0.0,
418
+ "norm_key_precision": 0.0,
419
+ "norm_key_recall": 0.0,
420
+ "norm_key_f1": 0.0
421
+ },
422
+ "tmf921_lifecycle_suspend": {
423
+ "num_examples": 3,
424
+ "parse_json": 0.6666666666666666,
425
+ "gold_parse_json": 1.0,
426
+ "exact_match": 0.0,
427
+ "field_precision": 0.0,
428
+ "field_recall": 0.0,
429
+ "field_f1": 0.0,
430
+ "field_tp": 0.0,
431
+ "field_fp": 2.3333333333333335,
432
+ "field_fn": 4.666666666666667,
433
+ "slice_sst_pass": 0.6666666666666666,
434
+ "kpi_text_presence_pass": 0.0,
435
+ "adversarial_status_pass": 0.6666666666666666,
436
+ "norm_parse_json": 0.6666666666666666,
437
+ "norm_gold_parse_json": 1.0,
438
+ "norm_exact_match": 0.0,
439
+ "norm_field_precision": 0.0,
440
+ "norm_field_recall": 0.0,
441
+ "norm_field_f1": 0.0,
442
+ "norm_key_precision": 0.0,
443
+ "norm_key_recall": 0.0,
444
+ "norm_key_f1": 0.0
445
+ }
446
+ },
447
+ "by_slice_type": {
448
+ "HMTC": {
449
+ "num_examples": 13,
450
+ "parse_json": 0.23076923076923078,
451
+ "gold_parse_json": 1.0,
452
+ "exact_match": 0.0,
453
+ "field_precision": 0.0,
454
+ "field_recall": 0.0,
455
+ "field_f1": 0.0,
456
+ "field_tp": 0.0,
457
+ "field_fp": 3.8461538461538463,
458
+ "field_fn": 5.461538461538462,
459
+ "slice_sst_pass": 0.23076923076923078,
460
+ "kpi_text_presence_pass": 0.23076923076923078,
461
+ "adversarial_status_pass": 0.23076923076923078,
462
+ "norm_parse_json": 0.23076923076923078,
463
+ "norm_gold_parse_json": 1.0,
464
+ "norm_exact_match": 0.0,
465
+ "norm_field_precision": 0.0,
466
+ "norm_field_recall": 0.0,
467
+ "norm_field_f1": 0.0,
468
+ "norm_key_precision": 0.0,
469
+ "norm_key_recall": 0.0,
470
+ "norm_key_f1": 0.0
471
+ },
472
+ "MPS": {
473
+ "num_examples": 28,
474
+ "parse_json": 0.2857142857142857,
475
+ "gold_parse_json": 1.0,
476
+ "exact_match": 0.0,
477
+ "field_precision": 0.005291005291005291,
478
+ "field_recall": 0.0035714285714285718,
479
+ "field_f1": 0.004264392324093817,
480
+ "field_tp": 0.14285714285714285,
481
+ "field_fp": 4.964285714285714,
482
+ "field_fn": 9.071428571428571,
483
+ "slice_sst_pass": 0.25,
484
+ "kpi_text_presence_pass": 0.25,
485
+ "adversarial_status_pass": 0.2857142857142857,
486
+ "norm_parse_json": 0.2857142857142857,
487
+ "norm_gold_parse_json": 1.0,
488
+ "norm_exact_match": 0.0,
489
+ "norm_field_precision": 0.0013227513227513227,
490
+ "norm_field_recall": 0.0009398496240601503,
491
+ "norm_field_f1": 0.001098901098901099,
492
+ "norm_field_tp": 0.125,
493
+ "norm_field_fp": 17.0,
494
+ "norm_field_fn": 29.125,
495
+ "norm_key_precision": 0.01912626912626913,
496
+ "norm_key_recall": 0.013268465280849183,
497
+ "norm_key_f1": 0.01564625850340136,
498
+ "norm_key_tp": 1.875,
499
+ "norm_key_fp": 15.25,
500
+ "norm_key_fn": 27.375
501
+ },
502
+ "URLLC": {
503
+ "num_examples": 53,
504
+ "parse_json": 0.39622641509433965,
505
+ "gold_parse_json": 1.0,
506
+ "exact_match": 0.0,
507
+ "field_precision": 0.0005896226415094339,
508
+ "field_recall": 0.0004716981132075472,
509
+ "field_f1": 0.0005241090146750524,
510
+ "field_tp": 0.018867924528301886,
511
+ "field_fp": 8.075471698113208,
512
+ "field_fn": 13.09433962264151,
513
+ "slice_sst_pass": 0.33962264150943394,
514
+ "kpi_text_presence_pass": 0.37735849056603776,
515
+ "adversarial_status_pass": 0.39622641509433965,
516
+ "norm_parse_json": 0.39622641509433965,
517
+ "norm_gold_parse_json": 1.0,
518
+ "norm_exact_match": 0.0,
519
+ "norm_field_precision": 0.001768867924528302,
520
+ "norm_field_recall": 0.0014895729890764646,
521
+ "norm_field_f1": 0.0016172506738544475,
522
+ "norm_key_precision": 0.03384419595155658,
523
+ "norm_key_recall": 0.018686839184531807,
524
+ "norm_key_f1": 0.022979510237634688
525
+ },
526
+ "V2X": {
527
+ "num_examples": 23,
528
+ "parse_json": 0.30434782608695654,
529
+ "gold_parse_json": 1.0,
530
+ "exact_match": 0.0,
531
+ "field_precision": 0.0,
532
+ "field_recall": 0.0,
533
+ "field_f1": 0.0,
534
+ "field_tp": 0.0,
535
+ "field_fp": 6.391304347826087,
536
+ "field_fn": 10.478260869565217,
537
+ "slice_sst_pass": 0.21739130434782608,
538
+ "kpi_text_presence_pass": 0.2608695652173913,
539
+ "adversarial_status_pass": 0.30434782608695654,
540
+ "norm_parse_json": 0.30434782608695654,
541
+ "norm_gold_parse_json": 1.0,
542
+ "norm_exact_match": 0.0,
543
+ "norm_field_precision": 0.0,
544
+ "norm_field_recall": 0.0,
545
+ "norm_field_f1": 0.0,
546
+ "norm_field_tp": 0.0,
547
+ "norm_field_fp": 18.285714285714285,
548
+ "norm_field_fn": 31.0,
549
+ "norm_key_precision": 0.0399021268586486,
550
+ "norm_key_recall": 0.01432225063938619,
551
+ "norm_key_f1": 0.02104558281915148,
552
+ "norm_key_tp": 1.1428571428571428,
553
+ "norm_key_fp": 17.142857142857142,
554
+ "norm_key_fn": 29.857142857142858
555
+ },
556
+ "eMBB": {
557
+ "num_examples": 62,
558
+ "parse_json": 0.3225806451612903,
559
+ "gold_parse_json": 1.0,
560
+ "exact_match": 0.0,
561
+ "field_precision": 0.0020161290322580645,
562
+ "field_recall": 0.0007012622720897616,
563
+ "field_f1": 0.0010405827263267429,
564
+ "field_tp": 0.016129032258064516,
565
+ "field_fp": 4.661290322580645,
566
+ "field_fn": 10.661290322580646,
567
+ "slice_sst_pass": 0.2903225806451613,
568
+ "kpi_text_presence_pass": 0.22580645161290322,
569
+ "adversarial_status_pass": 0.3225806451612903,
570
+ "norm_parse_json": 0.3225806451612903,
571
+ "norm_gold_parse_json": 1.0,
572
+ "norm_exact_match": 0.0,
573
+ "norm_field_precision": 0.0020161290322580645,
574
+ "norm_field_recall": 0.0008064516129032258,
575
+ "norm_field_f1": 0.0011520737327188942,
576
+ "norm_key_precision": 0.019946099056733495,
577
+ "norm_key_recall": 0.00824927594127634,
578
+ "norm_key_f1": 0.011556938739101928
579
+ },
580
+ "mMTC": {
581
+ "num_examples": 21,
582
+ "parse_json": 0.38095238095238093,
583
+ "gold_parse_json": 1.0,
584
+ "exact_match": 0.0,
585
+ "field_precision": 0.0,
586
+ "field_recall": 0.0,
587
+ "field_f1": 0.0,
588
+ "field_tp": 0.0,
589
+ "field_fp": 5.619047619047619,
590
+ "field_fn": 8.666666666666666,
591
+ "slice_sst_pass": 0.2857142857142857,
592
+ "kpi_text_presence_pass": 0.23809523809523808,
593
+ "adversarial_status_pass": 0.38095238095238093,
594
+ "norm_parse_json": 0.38095238095238093,
595
+ "norm_gold_parse_json": 1.0,
596
+ "norm_exact_match": 0.0,
597
+ "norm_field_precision": 0.0,
598
+ "norm_field_recall": 0.0,
599
+ "norm_field_f1": 0.0,
600
+ "norm_key_precision": 0.04325396825396825,
601
+ "norm_key_recall": 0.017919799498746863,
602
+ "norm_key_f1": 0.02521008403361344
603
+ }
604
+ },
605
+ "by_lifecycle_operation": {
606
+ "activate": {
607
+ "num_examples": 2,
608
+ "parse_json": 0.0,
609
+ "gold_parse_json": 1.0,
610
+ "exact_match": 0.0,
611
+ "field_precision": 0.0,
612
+ "field_recall": 0.0,
613
+ "field_f1": 0.0,
614
+ "field_tp": 0.0,
615
+ "field_fp": 0.0,
616
+ "field_fn": 0.0,
617
+ "slice_sst_pass": 0.0,
618
+ "kpi_text_presence_pass": 0.0,
619
+ "adversarial_status_pass": 0.0,
620
+ "norm_parse_json": 0.0,
621
+ "norm_gold_parse_json": 1.0,
622
+ "norm_exact_match": 0.0,
623
+ "norm_field_precision": 0.0,
624
+ "norm_field_recall": 0.0,
625
+ "norm_field_f1": 0.0,
626
+ "norm_key_precision": 0.0,
627
+ "norm_key_recall": 0.0,
628
+ "norm_key_f1": 0.0
629
+ },
630
+ "create": {
631
+ "num_examples": 186,
632
+ "parse_json": 0.3387096774193548,
633
+ "gold_parse_json": 1.0,
634
+ "exact_match": 0.0,
635
+ "field_precision": 0.0016365491835921943,
636
+ "field_recall": 0.0009057971014492754,
637
+ "field_f1": 0.0011381553327132405,
638
+ "field_tp": 0.03225806451612903,
639
+ "field_fp": 6.198924731182796,
640
+ "field_fn": 11.06989247311828,
641
+ "slice_sst_pass": 0.2849462365591398,
642
+ "kpi_text_presence_pass": 0.2956989247311828,
643
+ "adversarial_status_pass": 0.3387096774193548,
644
+ "norm_parse_json": 0.3387096774193548,
645
+ "norm_gold_parse_json": 1.0,
646
+ "norm_exact_match": 0.0,
647
+ "norm_field_precision": 0.001375199123855038,
648
+ "norm_field_recall": 0.0008347481607243915,
649
+ "norm_field_f1": 0.001010280042538107,
650
+ "norm_key_precision": 0.028989345771869686,
651
+ "norm_key_recall": 0.013866140683777756,
652
+ "norm_key_f1": 0.01820429920462668
653
+ },
654
+ "report": {
655
+ "num_examples": 4,
656
+ "parse_json": 0.0,
657
+ "gold_parse_json": 1.0,
658
+ "exact_match": 0.0,
659
+ "field_precision": 0.0,
660
+ "field_recall": 0.0,
661
+ "field_f1": 0.0,
662
+ "field_tp": 0.0,
663
+ "field_fp": 0.0,
664
+ "field_fn": 0.0,
665
+ "slice_sst_pass": 0.0,
666
+ "kpi_text_presence_pass": 0.0,
667
+ "adversarial_status_pass": 0.0,
668
+ "norm_parse_json": 0.0,
669
+ "norm_gold_parse_json": 1.0,
670
+ "norm_exact_match": 0.0,
671
+ "norm_field_precision": 0.0,
672
+ "norm_field_recall": 0.0,
673
+ "norm_field_f1": 0.0,
674
+ "norm_key_precision": 0.0,
675
+ "norm_key_recall": 0.0,
676
+ "norm_key_f1": 0.0
677
+ },
678
+ "resume": {
679
+ "num_examples": 2,
680
+ "parse_json": 0.0,
681
+ "gold_parse_json": 1.0,
682
+ "exact_match": 0.0,
683
+ "field_precision": 0.0,
684
+ "field_recall": 0.0,
685
+ "field_f1": 0.0,
686
+ "field_tp": 0.0,
687
+ "field_fp": 0.0,
688
+ "field_fn": 0.0,
689
+ "slice_sst_pass": 0.0,
690
+ "kpi_text_presence_pass": 0.0,
691
+ "adversarial_status_pass": 0.0,
692
+ "norm_parse_json": 0.0,
693
+ "norm_gold_parse_json": 1.0,
694
+ "norm_exact_match": 0.0,
695
+ "norm_field_precision": 0.0,
696
+ "norm_field_recall": 0.0,
697
+ "norm_field_f1": 0.0,
698
+ "norm_key_precision": 0.0,
699
+ "norm_key_recall": 0.0,
700
+ "norm_key_f1": 0.0
701
+ },
702
+ "scale": {
703
+ "num_examples": 3,
704
+ "parse_json": 0.6666666666666666,
705
+ "gold_parse_json": 1.0,
706
+ "exact_match": 0.0,
707
+ "field_precision": 0.0,
708
+ "field_recall": 0.0,
709
+ "field_f1": 0.0,
710
+ "field_tp": 0.0,
711
+ "field_fp": 3.6666666666666665,
712
+ "field_fn": 10.0,
713
+ "slice_sst_pass": 0.6666666666666666,
714
+ "kpi_text_presence_pass": 0.0,
715
+ "adversarial_status_pass": 0.6666666666666666,
716
+ "norm_parse_json": 0.6666666666666666,
717
+ "norm_gold_parse_json": 1.0,
718
+ "norm_exact_match": 0.0,
719
+ "norm_field_precision": 0.0,
720
+ "norm_field_recall": 0.0,
721
+ "norm_field_f1": 0.0,
722
+ "norm_key_precision": 0.0,
723
+ "norm_key_recall": 0.0,
724
+ "norm_key_f1": 0.0
725
+ },
726
+ "suspend": {
727
+ "num_examples": 3,
728
+ "parse_json": 0.6666666666666666,
729
+ "gold_parse_json": 1.0,
730
+ "exact_match": 0.0,
731
+ "field_precision": 0.0,
732
+ "field_recall": 0.0,
733
+ "field_f1": 0.0,
734
+ "field_tp": 0.0,
735
+ "field_fp": 2.3333333333333335,
736
+ "field_fn": 4.666666666666667,
737
+ "slice_sst_pass": 0.6666666666666666,
738
+ "kpi_text_presence_pass": 0.0,
739
+ "adversarial_status_pass": 0.6666666666666666,
740
+ "norm_parse_json": 0.6666666666666666,
741
+ "norm_gold_parse_json": 1.0,
742
+ "norm_exact_match": 0.0,
743
+ "norm_field_precision": 0.0,
744
+ "norm_field_recall": 0.0,
745
+ "norm_field_f1": 0.0,
746
+ "norm_key_precision": 0.0,
747
+ "norm_key_recall": 0.0,
748
+ "norm_key_f1": 0.0
749
+ }
750
+ }
751
+ },
752
+ "test_sector_ood": {
753
+ "num_examples": 200,
754
+ "parse_json": 0.345,
755
+ "gold_parse_json": 1.0,
756
+ "exact_match": 0.0,
757
+ "field_precision": 0.0012675865800865801,
758
+ "field_recall": 0.0008765822784810127,
759
+ "field_f1": 0.001016076991544281,
760
+ "field_tp": 0.04,
761
+ "field_fp": 7.095,
762
+ "field_fn": 12.085,
763
+ "slice_sst_pass": 0.28,
764
+ "kpi_text_presence_pass": 0.315,
765
+ "adversarial_status_pass": 0.345,
766
+ "norm_parse_json": 0.345,
767
+ "norm_gold_parse_json": 1.0,
768
+ "norm_exact_match": 0.0,
769
+ "norm_field_precision": 0.0010274343018908236,
770
+ "norm_field_recall": 0.000673374613003096,
771
+ "norm_field_f1": 0.0007733821733821733,
772
+ "norm_key_precision": 0.026291418303017844,
773
+ "norm_key_recall": 0.013494582043343653,
774
+ "norm_key_f1": 0.017133099321501633,
775
+ "by_target_layer": {
776
+ "a1_policy": {
777
+ "num_examples": 29,
778
+ "parse_json": 0.13793103448275862,
779
+ "gold_parse_json": 1.0,
780
+ "exact_match": 0.0,
781
+ "field_precision": 0.0,
782
+ "field_recall": 0.0,
783
+ "field_f1": 0.0,
784
+ "field_tp": 0.0,
785
+ "field_fp": 1.6206896551724137,
786
+ "field_fn": 2.6206896551724137,
787
+ "slice_sst_pass": 0.10344827586206896,
788
+ "kpi_text_presence_pass": 0.13793103448275862,
789
+ "adversarial_status_pass": 0.13793103448275862,
790
+ "norm_parse_json": 0.13793103448275862,
791
+ "norm_gold_parse_json": 1.0,
792
+ "norm_exact_match": 0.0,
793
+ "norm_field_precision": 0.0,
794
+ "norm_field_recall": 0.0,
795
+ "norm_field_f1": 0.0,
796
+ "norm_key_precision": 0.0,
797
+ "norm_key_recall": 0.0,
798
+ "norm_key_f1": 0.0
799
+ },
800
+ "camara": {
801
+ "num_examples": 43,
802
+ "parse_json": 0.6046511627906976,
803
+ "gold_parse_json": 1.0,
804
+ "exact_match": 0.0,
805
+ "field_precision": 0.0,
806
+ "field_recall": 0.0,
807
+ "field_f1": 0.0,
808
+ "field_tp": 0.0,
809
+ "field_fp": 4.837209302325581,
810
+ "field_fn": 9.744186046511627,
811
+ "slice_sst_pass": 0.5116279069767442,
812
+ "kpi_text_presence_pass": 0.6046511627906976,
813
+ "adversarial_status_pass": 0.6046511627906976,
814
+ "norm_parse_json": 0.6046511627906976,
815
+ "norm_gold_parse_json": 1.0,
816
+ "norm_exact_match": 0.0,
817
+ "norm_field_precision": 0.0,
818
+ "norm_field_recall": 0.0,
819
+ "norm_field_f1": 0.0,
820
+ "norm_field_tp": 0.0,
821
+ "norm_field_fp": 8.0,
822
+ "norm_field_fn": 14.115384615384615,
823
+ "norm_key_precision": 0.0,
824
+ "norm_key_recall": 0.0,
825
+ "norm_key_f1": 0.0,
826
+ "norm_key_tp": 0.0,
827
+ "norm_key_fp": 8.0,
828
+ "norm_key_fn": 14.115384615384615
829
+ },
830
+ "etsi_zsm": {
831
+ "num_examples": 17,
832
+ "parse_json": 0.5882352941176471,
833
+ "gold_parse_json": 1.0,
834
+ "exact_match": 0.0,
835
+ "field_precision": 0.0,
836
+ "field_recall": 0.0,
837
+ "field_f1": 0.0,
838
+ "field_tp": 0.0,
839
+ "field_fp": 20.941176470588236,
840
+ "field_fn": 33.529411764705884,
841
+ "slice_sst_pass": 0.5294117647058824,
842
+ "kpi_text_presence_pass": 0.5882352941176471,
843
+ "adversarial_status_pass": 0.5882352941176471,
844
+ "norm_parse_json": 0.5882352941176471,
845
+ "norm_gold_parse_json": 1.0,
846
+ "norm_exact_match": 0.0,
847
+ "norm_field_precision": 0.0,
848
+ "norm_field_recall": 0.0,
849
+ "norm_field_f1": 0.0,
850
+ "norm_field_tp": 0.0,
851
+ "norm_field_fp": 32.2,
852
+ "norm_field_fn": 55.0,
853
+ "norm_key_precision": 0.0,
854
+ "norm_key_recall": 0.0,
855
+ "norm_key_f1": 0.0,
856
+ "norm_key_tp": 0.0,
857
+ "norm_key_fp": 32.2,
858
+ "norm_key_fn": 55.0
859
+ },
860
+ "intent_3gpp": {
861
+ "num_examples": 34,
862
+ "parse_json": 0.4411764705882353,
863
+ "gold_parse_json": 1.0,
864
+ "exact_match": 0.0,
865
+ "field_precision": 0.0055147058823529415,
866
+ "field_recall": 0.004411764705882353,
867
+ "field_f1": 0.0049019607843137246,
868
+ "field_tp": 0.17647058823529413,
869
+ "field_fp": 14.823529411764707,
870
+ "field_fn": 17.470588235294116,
871
+ "slice_sst_pass": 0.3235294117647059,
872
+ "kpi_text_presence_pass": 0.4411764705882353,
873
+ "adversarial_status_pass": 0.4411764705882353,
874
+ "norm_parse_json": 0.4411764705882353,
875
+ "norm_gold_parse_json": 1.0,
876
+ "norm_exact_match": 0.0,
877
+ "norm_field_precision": 0.0034280604133545313,
878
+ "norm_field_recall": 0.003095975232198142,
879
+ "norm_field_f1": 0.0032492997198879554,
880
+ "norm_key_precision": 0.038850307236954165,
881
+ "norm_key_recall": 0.0348297213622291,
882
+ "norm_key_f1": 0.0366445612417816
883
+ },
884
+ "o1_nrm": {
885
+ "num_examples": 21,
886
+ "parse_json": 0.19047619047619047,
887
+ "gold_parse_json": 1.0,
888
+ "exact_match": 0.0,
889
+ "field_precision": 0.0,
890
+ "field_recall": 0.0,
891
+ "field_f1": 0.0,
892
+ "field_tp": 0.0,
893
+ "field_fp": 1.8095238095238095,
894
+ "field_fn": 4.380952380952381,
895
+ "slice_sst_pass": 0.14285714285714285,
896
+ "kpi_text_presence_pass": 0.0,
897
+ "adversarial_status_pass": 0.19047619047619047,
898
+ "norm_parse_json": 0.19047619047619047,
899
+ "norm_gold_parse_json": 1.0,
900
+ "norm_exact_match": 0.0,
901
+ "norm_field_precision": 0.0,
902
+ "norm_field_recall": 0.0,
903
+ "norm_field_f1": 0.0,
904
+ "norm_key_precision": 0.14928193499622072,
905
+ "norm_key_recall": 0.05952380952380952,
906
+ "norm_key_f1": 0.08496885344175983
907
+ },
908
+ "tmf921": {
909
+ "num_examples": 51,
910
+ "parse_json": 0.1568627450980392,
911
+ "gold_parse_json": 1.0,
912
+ "exact_match": 0.0,
913
+ "field_precision": 0.001294457176810118,
914
+ "field_recall": 0.0004964010920824026,
915
+ "field_f1": 0.0007166417969056782,
916
+ "field_tp": 0.0392156862745098,
917
+ "field_fp": 4.647058823529412,
918
+ "field_fn": 12.352941176470589,
919
+ "slice_sst_pass": 0.13725490196078433,
920
+ "kpi_text_presence_pass": 0.1568627450980392,
921
+ "adversarial_status_pass": 0.1568627450980392,
922
+ "norm_parse_json": 0.1568627450980392,
923
+ "norm_gold_parse_json": 1.0,
924
+ "norm_exact_match": 0.0,
925
+ "norm_field_precision": 0.0017437805161590327,
926
+ "norm_field_recall": 0.0005767012687427913,
927
+ "norm_field_f1": 0.0008666714549067489,
928
+ "norm_key_precision": 0.015734364306401818,
929
+ "norm_key_recall": 0.005190311418685121,
930
+ "norm_key_f1": 0.007771742349074426
931
+ },
932
+ "tmf921_lifecycle_monitor": {
933
+ "num_examples": 3,
934
+ "parse_json": 0.3333333333333333,
935
+ "gold_parse_json": 1.0,
936
+ "exact_match": 0.0,
937
+ "field_precision": 0.0,
938
+ "field_recall": 0.0,
939
+ "field_f1": 0.0,
940
+ "field_tp": 0.0,
941
+ "field_fp": 8.666666666666666,
942
+ "field_fn": 10.0,
943
+ "slice_sst_pass": 0.3333333333333333,
944
+ "kpi_text_presence_pass": 0.0,
945
+ "adversarial_status_pass": 0.3333333333333333,
946
+ "norm_parse_json": 0.3333333333333333,
947
+ "norm_gold_parse_json": 1.0,
948
+ "norm_exact_match": 0.0,
949
+ "norm_field_precision": 0.0,
950
+ "norm_field_recall": 0.0,
951
+ "norm_field_f1": 0.0,
952
+ "norm_field_tp": 0.0,
953
+ "norm_field_fp": 21.0,
954
+ "norm_field_fn": 29.0,
955
+ "norm_key_precision": 0.0,
956
+ "norm_key_recall": 0.0,
957
+ "norm_key_f1": 0.0,
958
+ "norm_key_tp": 0.0,
959
+ "norm_key_fp": 21.0,
960
+ "norm_key_fn": 29.0
961
+ },
962
+ "tmf921_lifecycle_report": {
963
+ "num_examples": 1,
964
+ "parse_json": 0.0,
965
+ "gold_parse_json": 1.0,
966
+ "exact_match": 0.0,
967
+ "field_precision": 0.0,
968
+ "field_recall": 0.0,
969
+ "field_f1": 0.0,
970
+ "field_tp": 0.0,
971
+ "field_fp": 0.0,
972
+ "field_fn": 0.0,
973
+ "slice_sst_pass": 0.0,
974
+ "kpi_text_presence_pass": 0.0,
975
+ "adversarial_status_pass": 0.0,
976
+ "norm_parse_json": 0.0,
977
+ "norm_gold_parse_json": 1.0,
978
+ "norm_exact_match": 0.0,
979
+ "norm_field_precision": 0.0,
980
+ "norm_field_recall": 0.0,
981
+ "norm_field_f1": 0.0,
982
+ "norm_key_precision": 0.0,
983
+ "norm_key_recall": 0.0,
984
+ "norm_key_f1": 0.0
985
+ },
986
+ "tmf921_lifecycle_resume": {
987
+ "num_examples": 1,
988
+ "parse_json": 1.0,
989
+ "gold_parse_json": 1.0,
990
+ "exact_match": 0.0,
991
+ "field_precision": 0.0,
992
+ "field_recall": 0.0,
993
+ "field_f1": 0.0,
994
+ "field_tp": 0.0,
995
+ "field_fp": 3.0,
996
+ "field_fn": 6.0,
997
+ "slice_sst_pass": 0.0,
998
+ "kpi_text_presence_pass": 0.0,
999
+ "adversarial_status_pass": 1.0,
1000
+ "norm_parse_json": 1.0,
1001
+ "norm_gold_parse_json": 1.0,
1002
+ "norm_exact_match": 0.0,
1003
+ "norm_field_precision": 0.0,
1004
+ "norm_field_recall": 0.0,
1005
+ "norm_field_f1": 0.0,
1006
+ "norm_field_tp": 0.0,
1007
+ "norm_field_fp": 2.0,
1008
+ "norm_field_fn": 5.0,
1009
+ "norm_key_precision": 0.0,
1010
+ "norm_key_recall": 0.0,
1011
+ "norm_key_f1": 0.0,
1012
+ "norm_key_tp": 0.0,
1013
+ "norm_key_fp": 2.0,
1014
+ "norm_key_fn": 5.0
1015
+ }
1016
+ },
1017
+ "by_slice_type": {
1018
+ "HMTC": {
1019
+ "num_examples": 17,
1020
+ "parse_json": 0.4117647058823529,
1021
+ "gold_parse_json": 1.0,
1022
+ "exact_match": 0.0,
1023
+ "field_precision": 0.0,
1024
+ "field_recall": 0.0,
1025
+ "field_f1": 0.0,
1026
+ "field_tp": 0.0,
1027
+ "field_fp": 8.176470588235293,
1028
+ "field_fn": 13.352941176470589,
1029
+ "slice_sst_pass": 0.35294117647058826,
1030
+ "kpi_text_presence_pass": 0.35294117647058826,
1031
+ "adversarial_status_pass": 0.4117647058823529,
1032
+ "norm_parse_json": 0.4117647058823529,
1033
+ "norm_gold_parse_json": 1.0,
1034
+ "norm_exact_match": 0.0,
1035
+ "norm_field_precision": 0.003179650238473768,
1036
+ "norm_field_recall": 0.003095975232198142,
1037
+ "norm_field_f1": 0.0031372549019607846,
1038
+ "norm_field_tp": 0.2857142857142857,
1039
+ "norm_field_fp": 17.857142857142858,
1040
+ "norm_field_fn": 28.714285714285715,
1041
+ "norm_key_precision": 0.08990570945030148,
1042
+ "norm_key_recall": 0.053432890183937355,
1043
+ "norm_key_f1": 0.06324097755044046,
1044
+ "norm_key_tp": 4.285714285714286,
1045
+ "norm_key_fp": 13.857142857142858,
1046
+ "norm_key_fn": 24.714285714285715
1047
+ },
1048
+ "MPS": {
1049
+ "num_examples": 19,
1050
+ "parse_json": 0.2631578947368421,
1051
+ "gold_parse_json": 1.0,
1052
+ "exact_match": 0.0,
1053
+ "field_precision": 0.0,
1054
+ "field_recall": 0.0,
1055
+ "field_f1": 0.0,
1056
+ "field_tp": 0.0,
1057
+ "field_fp": 4.7368421052631575,
1058
+ "field_fn": 6.947368421052632,
1059
+ "slice_sst_pass": 0.2631578947368421,
1060
+ "kpi_text_presence_pass": 0.2631578947368421,
1061
+ "adversarial_status_pass": 0.2631578947368421,
1062
+ "norm_parse_json": 0.2631578947368421,
1063
+ "norm_gold_parse_json": 1.0,
1064
+ "norm_exact_match": 0.0,
1065
+ "norm_field_precision": 0.0,
1066
+ "norm_field_recall": 0.0,
1067
+ "norm_field_f1": 0.0,
1068
+ "norm_key_precision": 0.0018148820326678765,
1069
+ "norm_key_recall": 0.0013850415512465374,
1070
+ "norm_key_f1": 0.0015710919088766692
1071
+ },
1072
+ "URLLC": {
1073
+ "num_examples": 58,
1074
+ "parse_json": 0.3275862068965517,
1075
+ "gold_parse_json": 1.0,
1076
+ "exact_match": 0.0,
1077
+ "field_precision": 0.0,
1078
+ "field_recall": 0.0,
1079
+ "field_f1": 0.0,
1080
+ "field_tp": 0.0,
1081
+ "field_fp": 7.362068965517241,
1082
+ "field_fn": 12.293103448275861,
1083
+ "slice_sst_pass": 0.25862068965517243,
1084
+ "kpi_text_presence_pass": 0.29310344827586204,
1085
+ "adversarial_status_pass": 0.3275862068965517,
1086
+ "norm_parse_json": 0.3275862068965517,
1087
+ "norm_gold_parse_json": 1.0,
1088
+ "norm_exact_match": 0.0,
1089
+ "norm_field_precision": 0.0,
1090
+ "norm_field_recall": 0.0,
1091
+ "norm_field_f1": 0.0,
1092
+ "norm_key_precision": 0.016931312620967795,
1093
+ "norm_key_recall": 0.007094053592398847,
1094
+ "norm_key_f1": 0.009892558559443798
1095
+ },
1096
+ "V2X": {
1097
+ "num_examples": 19,
1098
+ "parse_json": 0.21052631578947367,
1099
+ "gold_parse_json": 1.0,
1100
+ "exact_match": 0.0,
1101
+ "field_precision": 0.0,
1102
+ "field_recall": 0.0,
1103
+ "field_f1": 0.0,
1104
+ "field_tp": 0.0,
1105
+ "field_fp": 2.526315789473684,
1106
+ "field_fn": 4.421052631578948,
1107
+ "slice_sst_pass": 0.05263157894736842,
1108
+ "kpi_text_presence_pass": 0.10526315789473684,
1109
+ "adversarial_status_pass": 0.21052631578947367,
1110
+ "norm_parse_json": 0.21052631578947367,
1111
+ "norm_gold_parse_json": 1.0,
1112
+ "norm_exact_match": 0.0,
1113
+ "norm_field_precision": 0.0,
1114
+ "norm_field_recall": 0.0,
1115
+ "norm_field_f1": 0.0,
1116
+ "norm_key_precision": 0.041423001949317736,
1117
+ "norm_key_recall": 0.017174515235457065,
1118
+ "norm_key_f1": 0.024175824175824173
1119
+ },
1120
+ "eMBB": {
1121
+ "num_examples": 64,
1122
+ "parse_json": 0.4375,
1123
+ "gold_parse_json": 1.0,
1124
+ "exact_match": 0.0,
1125
+ "field_precision": 0.003961208062770563,
1126
+ "field_recall": 0.0027393196202531644,
1127
+ "field_f1": 0.0031752405985758783,
1128
+ "field_tp": 0.125,
1129
+ "field_fp": 8.59375,
1130
+ "field_fn": 15.1875,
1131
+ "slice_sst_pass": 0.421875,
1132
+ "kpi_text_presence_pass": 0.421875,
1133
+ "adversarial_status_pass": 0.4375,
1134
+ "norm_parse_json": 0.4375,
1135
+ "norm_gold_parse_json": 1.0,
1136
+ "norm_exact_match": 0.0,
1137
+ "norm_field_precision": 0.002366137598814229,
1138
+ "norm_field_recall": 0.0012819272445820434,
1139
+ "norm_field_f1": 0.0015834859584859585,
1140
+ "norm_key_precision": 0.026443379895316036,
1141
+ "norm_key_recall": 0.01470830108359133,
1142
+ "norm_key_f1": 0.018304564209736623
1143
+ },
1144
+ "mMTC": {
1145
+ "num_examples": 23,
1146
+ "parse_json": 0.2608695652173913,
1147
+ "gold_parse_json": 1.0,
1148
+ "exact_match": 0.0,
1149
+ "field_precision": 0.0,
1150
+ "field_recall": 0.0,
1151
+ "field_f1": 0.0,
1152
+ "field_tp": 0.0,
1153
+ "field_fp": 7.173913043478261,
1154
+ "field_fn": 12.565217391304348,
1155
+ "slice_sst_pass": 0.08695652173913043,
1156
+ "kpi_text_presence_pass": 0.2608695652173913,
1157
+ "adversarial_status_pass": 0.2608695652173913,
1158
+ "norm_parse_json": 0.2608695652173913,
1159
+ "norm_gold_parse_json": 1.0,
1160
+ "norm_exact_match": 0.0,
1161
+ "norm_field_precision": 0.0,
1162
+ "norm_field_recall": 0.0,
1163
+ "norm_field_f1": 0.0,
1164
+ "norm_key_precision": 0.010172798216276478,
1165
+ "norm_key_recall": 0.0037017095167586485,
1166
+ "norm_key_f1": 0.005090058020114731
1167
+ }
1168
+ },
1169
+ "by_lifecycle_operation": {
1170
+ "create": {
1171
+ "num_examples": 195,
1172
+ "parse_json": 0.3435897435897436,
1173
+ "gold_parse_json": 1.0,
1174
+ "exact_match": 0.0,
1175
+ "field_precision": 0.0013000888000888001,
1176
+ "field_recall": 0.0008990587471600129,
1177
+ "field_f1": 0.001042130247737724,
1178
+ "field_tp": 0.041025641025641026,
1179
+ "field_fp": 7.128205128205129,
1180
+ "field_fn": 12.21025641025641,
1181
+ "slice_sst_pass": 0.28205128205128205,
1182
+ "kpi_text_presence_pass": 0.3230769230769231,
1183
+ "adversarial_status_pass": 0.3435897435897436,
1184
+ "norm_parse_json": 0.3435897435897436,
1185
+ "norm_gold_parse_json": 1.0,
1186
+ "norm_exact_match": 0.0,
1187
+ "norm_field_precision": 0.0010537787711700755,
1188
+ "norm_field_recall": 0.0006906406287211241,
1189
+ "norm_field_f1": 0.0007932124855201777,
1190
+ "norm_key_precision": 0.026965557233864457,
1191
+ "norm_key_recall": 0.01384059696753195,
1192
+ "norm_key_f1": 0.017572409560514497
1193
+ },
1194
+ "monitor": {
1195
+ "num_examples": 3,
1196
+ "parse_json": 0.3333333333333333,
1197
+ "gold_parse_json": 1.0,
1198
+ "exact_match": 0.0,
1199
+ "field_precision": 0.0,
1200
+ "field_recall": 0.0,
1201
+ "field_f1": 0.0,
1202
+ "field_tp": 0.0,
1203
+ "field_fp": 8.666666666666666,
1204
+ "field_fn": 10.0,
1205
+ "slice_sst_pass": 0.3333333333333333,
1206
+ "kpi_text_presence_pass": 0.0,
1207
+ "adversarial_status_pass": 0.3333333333333333,
1208
+ "norm_parse_json": 0.3333333333333333,
1209
+ "norm_gold_parse_json": 1.0,
1210
+ "norm_exact_match": 0.0,
1211
+ "norm_field_precision": 0.0,
1212
+ "norm_field_recall": 0.0,
1213
+ "norm_field_f1": 0.0,
1214
+ "norm_field_tp": 0.0,
1215
+ "norm_field_fp": 21.0,
1216
+ "norm_field_fn": 29.0,
1217
+ "norm_key_precision": 0.0,
1218
+ "norm_key_recall": 0.0,
1219
+ "norm_key_f1": 0.0,
1220
+ "norm_key_tp": 0.0,
1221
+ "norm_key_fp": 21.0,
1222
+ "norm_key_fn": 29.0
1223
+ },
1224
+ "report": {
1225
+ "num_examples": 1,
1226
+ "parse_json": 0.0,
1227
+ "gold_parse_json": 1.0,
1228
+ "exact_match": 0.0,
1229
+ "field_precision": 0.0,
1230
+ "field_recall": 0.0,
1231
+ "field_f1": 0.0,
1232
+ "field_tp": 0.0,
1233
+ "field_fp": 0.0,
1234
+ "field_fn": 0.0,
1235
+ "slice_sst_pass": 0.0,
1236
+ "kpi_text_presence_pass": 0.0,
1237
+ "adversarial_status_pass": 0.0,
1238
+ "norm_parse_json": 0.0,
1239
+ "norm_gold_parse_json": 1.0,
1240
+ "norm_exact_match": 0.0,
1241
+ "norm_field_precision": 0.0,
1242
+ "norm_field_recall": 0.0,
1243
+ "norm_field_f1": 0.0,
1244
+ "norm_key_precision": 0.0,
1245
+ "norm_key_recall": 0.0,
1246
+ "norm_key_f1": 0.0
1247
+ },
1248
+ "resume": {
1249
+ "num_examples": 1,
1250
+ "parse_json": 1.0,
1251
+ "gold_parse_json": 1.0,
1252
+ "exact_match": 0.0,
1253
+ "field_precision": 0.0,
1254
+ "field_recall": 0.0,
1255
+ "field_f1": 0.0,
1256
+ "field_tp": 0.0,
1257
+ "field_fp": 3.0,
1258
+ "field_fn": 6.0,
1259
+ "slice_sst_pass": 0.0,
1260
+ "kpi_text_presence_pass": 0.0,
1261
+ "adversarial_status_pass": 1.0,
1262
+ "norm_parse_json": 1.0,
1263
+ "norm_gold_parse_json": 1.0,
1264
+ "norm_exact_match": 0.0,
1265
+ "norm_field_precision": 0.0,
1266
+ "norm_field_recall": 0.0,
1267
+ "norm_field_f1": 0.0,
1268
+ "norm_field_tp": 0.0,
1269
+ "norm_field_fp": 2.0,
1270
+ "norm_field_fn": 5.0,
1271
+ "norm_key_precision": 0.0,
1272
+ "norm_key_recall": 0.0,
1273
+ "norm_key_f1": 0.0,
1274
+ "norm_key_tp": 0.0,
1275
+ "norm_key_fp": 2.0,
1276
+ "norm_key_fn": 5.0
1277
+ }
1278
+ }
1279
+ },
1280
+ "test_template_ood": {
1281
+ "num_examples": 200,
1282
+ "parse_json": 0.34,
1283
+ "gold_parse_json": 1.0,
1284
+ "exact_match": 0.0,
1285
+ "field_precision": 0.0010798239750445633,
1286
+ "field_recall": 0.0008132911392405063,
1287
+ "field_f1": 0.0009188652938652939,
1288
+ "field_tp": 0.035,
1289
+ "field_fp": 7.145,
1290
+ "field_fn": 11.085,
1291
+ "slice_sst_pass": 0.21,
1292
+ "kpi_text_presence_pass": 0.3,
1293
+ "adversarial_status_pass": 0.34,
1294
+ "norm_parse_json": 0.34,
1295
+ "norm_gold_parse_json": 1.0,
1296
+ "norm_exact_match": 0.0,
1297
+ "norm_field_precision": 0.0017085360459973462,
1298
+ "norm_field_recall": 0.0012577399380804951,
1299
+ "norm_field_f1": 0.001413544991721657,
1300
+ "norm_key_precision": 0.020762022861588486,
1301
+ "norm_key_recall": 0.015035603715170279,
1302
+ "norm_key_f1": 0.017150005183047948,
1303
+ "by_target_layer": {
1304
+ "a1_policy": {
1305
+ "num_examples": 33,
1306
+ "parse_json": 0.2727272727272727,
1307
+ "gold_parse_json": 1.0,
1308
+ "exact_match": 0.0,
1309
+ "field_precision": 0.0,
1310
+ "field_recall": 0.0,
1311
+ "field_f1": 0.0,
1312
+ "field_tp": 0.0,
1313
+ "field_fp": 3.3333333333333335,
1314
+ "field_fn": 5.181818181818182,
1315
+ "slice_sst_pass": 0.24242424242424243,
1316
+ "kpi_text_presence_pass": 0.21212121212121213,
1317
+ "adversarial_status_pass": 0.2727272727272727,
1318
+ "norm_parse_json": 0.2727272727272727,
1319
+ "norm_gold_parse_json": 1.0,
1320
+ "norm_exact_match": 0.0,
1321
+ "norm_field_precision": 0.0,
1322
+ "norm_field_recall": 0.0,
1323
+ "norm_field_f1": 0.0,
1324
+ "norm_key_precision": 0.0,
1325
+ "norm_key_recall": 0.0,
1326
+ "norm_key_f1": 0.0
1327
+ },
1328
+ "camara": {
1329
+ "num_examples": 29,
1330
+ "parse_json": 0.7241379310344828,
1331
+ "gold_parse_json": 1.0,
1332
+ "exact_match": 0.0,
1333
+ "field_precision": 0.0,
1334
+ "field_recall": 0.0,
1335
+ "field_f1": 0.0,
1336
+ "field_tp": 0.0,
1337
+ "field_fp": 5.862068965517241,
1338
+ "field_fn": 11.482758620689655,
1339
+ "slice_sst_pass": 0.41379310344827586,
1340
+ "kpi_text_presence_pass": 0.7241379310344828,
1341
+ "adversarial_status_pass": 0.7241379310344828,
1342
+ "norm_parse_json": 0.7241379310344828,
1343
+ "norm_gold_parse_json": 1.0,
1344
+ "norm_exact_match": 0.0,
1345
+ "norm_field_precision": 0.0,
1346
+ "norm_field_recall": 0.0,
1347
+ "norm_field_f1": 0.0,
1348
+ "norm_field_tp": 0.0,
1349
+ "norm_field_fp": 8.095238095238095,
1350
+ "norm_field_fn": 13.857142857142858,
1351
+ "norm_key_precision": 0.0,
1352
+ "norm_key_recall": 0.0,
1353
+ "norm_key_f1": 0.0,
1354
+ "norm_key_tp": 0.0,
1355
+ "norm_key_fp": 8.095238095238095,
1356
+ "norm_key_fn": 13.857142857142858
1357
+ },
1358
+ "etsi_zsm": {
1359
+ "num_examples": 21,
1360
+ "parse_json": 0.47619047619047616,
1361
+ "gold_parse_json": 1.0,
1362
+ "exact_match": 0.0,
1363
+ "field_precision": 0.0,
1364
+ "field_recall": 0.0,
1365
+ "field_f1": 0.0,
1366
+ "field_tp": 0.0,
1367
+ "field_fp": 18.714285714285715,
1368
+ "field_fn": 27.142857142857142,
1369
+ "slice_sst_pass": 0.3333333333333333,
1370
+ "kpi_text_presence_pass": 0.47619047619047616,
1371
+ "adversarial_status_pass": 0.47619047619047616,
1372
+ "norm_parse_json": 0.47619047619047616,
1373
+ "norm_gold_parse_json": 1.0,
1374
+ "norm_exact_match": 0.0,
1375
+ "norm_field_precision": 0.0,
1376
+ "norm_field_recall": 0.0,
1377
+ "norm_field_f1": 0.0,
1378
+ "norm_key_precision": 0.0,
1379
+ "norm_key_recall": 0.0,
1380
+ "norm_key_f1": 0.0
1381
+ },
1382
+ "intent_3gpp": {
1383
+ "num_examples": 50,
1384
+ "parse_json": 0.38,
1385
+ "gold_parse_json": 1.0,
1386
+ "exact_match": 0.0,
1387
+ "field_precision": 0.0037132352941176474,
1388
+ "field_recall": 0.003,
1389
+ "field_f1": 0.0033183183183183185,
1390
+ "field_tp": 0.12,
1391
+ "field_fp": 11.62,
1392
+ "field_fn": 15.08,
1393
+ "slice_sst_pass": 0.18,
1394
+ "kpi_text_presence_pass": 0.36,
1395
+ "adversarial_status_pass": 0.38,
1396
+ "norm_parse_json": 0.38,
1397
+ "norm_gold_parse_json": 1.0,
1398
+ "norm_exact_match": 0.0,
1399
+ "norm_field_precision": 0.005781512605042016,
1400
+ "norm_field_recall": 0.004736842105263157,
1401
+ "norm_field_f1": 0.005194409851944098,
1402
+ "norm_key_precision": 0.054422553072531805,
1403
+ "norm_key_recall": 0.04578947368421052,
1404
+ "norm_key_f1": 0.04963327812912813
1405
+ },
1406
+ "o1_nrm": {
1407
+ "num_examples": 15,
1408
+ "parse_json": 0.13333333333333333,
1409
+ "gold_parse_json": 1.0,
1410
+ "exact_match": 0.0,
1411
+ "field_precision": 0.0,
1412
+ "field_recall": 0.0,
1413
+ "field_f1": 0.0,
1414
+ "field_tp": 0.0,
1415
+ "field_fp": 2.0,
1416
+ "field_fn": 3.066666666666667,
1417
+ "slice_sst_pass": 0.06666666666666667,
1418
+ "kpi_text_presence_pass": 0.0,
1419
+ "adversarial_status_pass": 0.13333333333333333,
1420
+ "norm_parse_json": 0.13333333333333333,
1421
+ "norm_gold_parse_json": 1.0,
1422
+ "norm_exact_match": 0.0,
1423
+ "norm_field_precision": 0.0,
1424
+ "norm_field_recall": 0.0,
1425
+ "norm_field_f1": 0.0,
1426
+ "norm_key_precision": 0.07272727272727272,
1427
+ "norm_key_recall": 0.04,
1428
+ "norm_key_f1": 0.051612903225806445
1429
+ },
1430
+ "tmf921": {
1431
+ "num_examples": 48,
1432
+ "parse_json": 0.08333333333333333,
1433
+ "gold_parse_json": 1.0,
1434
+ "exact_match": 0.0,
1435
+ "field_precision": 0.0006313131313131314,
1436
+ "field_recall": 0.00026371308016877635,
1437
+ "field_f1": 0.0003720238095238095,
1438
+ "field_tp": 0.020833333333333332,
1439
+ "field_fp": 2.7916666666666665,
1440
+ "field_fn": 6.5625,
1441
+ "slice_sst_pass": 0.0625,
1442
+ "kpi_text_presence_pass": 0.08333333333333333,
1443
+ "adversarial_status_pass": 0.08333333333333333,
1444
+ "norm_parse_json": 0.08333333333333333,
1445
+ "norm_gold_parse_json": 1.0,
1446
+ "norm_exact_match": 0.0,
1447
+ "norm_field_precision": 0.0010964912280701754,
1448
+ "norm_field_recall": 0.00030637254901960784,
1449
+ "norm_field_f1": 0.0004789272030651341,
1450
+ "norm_key_precision": 0.007090996412125321,
1451
+ "norm_key_recall": 0.0024509803921568627,
1452
+ "norm_key_f1": 0.0036279912867934645
1453
+ },
1454
+ "tmf921_lifecycle_resume": {
1455
+ "num_examples": 2,
1456
+ "parse_json": 0.5,
1457
+ "gold_parse_json": 1.0,
1458
+ "exact_match": 0.0,
1459
+ "field_precision": 0.0,
1460
+ "field_recall": 0.0,
1461
+ "field_f1": 0.0,
1462
+ "field_tp": 0.0,
1463
+ "field_fp": 1.5,
1464
+ "field_fn": 3.0,
1465
+ "slice_sst_pass": 0.5,
1466
+ "kpi_text_presence_pass": 0.0,
1467
+ "adversarial_status_pass": 0.5,
1468
+ "norm_parse_json": 0.5,
1469
+ "norm_gold_parse_json": 1.0,
1470
+ "norm_exact_match": 0.0,
1471
+ "norm_field_precision": 0.0,
1472
+ "norm_field_recall": 0.0,
1473
+ "norm_field_f1": 0.0,
1474
+ "norm_key_precision": 0.0,
1475
+ "norm_key_recall": 0.0,
1476
+ "norm_key_f1": 0.0
1477
+ },
1478
+ "tmf921_lifecycle_scale": {
1479
+ "num_examples": 1,
1480
+ "parse_json": 1.0,
1481
+ "gold_parse_json": 1.0,
1482
+ "exact_match": 0.0,
1483
+ "field_precision": 0.0,
1484
+ "field_recall": 0.0,
1485
+ "field_f1": 0.0,
1486
+ "field_tp": 0.0,
1487
+ "field_fp": 5.0,
1488
+ "field_fn": 15.0,
1489
+ "slice_sst_pass": 1.0,
1490
+ "kpi_text_presence_pass": 0.0,
1491
+ "adversarial_status_pass": 1.0,
1492
+ "norm_parse_json": 1.0,
1493
+ "norm_gold_parse_json": 1.0,
1494
+ "norm_exact_match": 0.0,
1495
+ "norm_field_precision": 0.0,
1496
+ "norm_field_recall": 0.0,
1497
+ "norm_field_f1": 0.0,
1498
+ "norm_field_tp": 0.0,
1499
+ "norm_field_fp": 5.0,
1500
+ "norm_field_fn": 13.0,
1501
+ "norm_key_precision": 0.0,
1502
+ "norm_key_recall": 0.0,
1503
+ "norm_key_f1": 0.0,
1504
+ "norm_key_tp": 0.0,
1505
+ "norm_key_fp": 5.0,
1506
+ "norm_key_fn": 13.0
1507
+ },
1508
+ "tmf921_lifecycle_suspend": {
1509
+ "num_examples": 1,
1510
+ "parse_json": 1.0,
1511
+ "gold_parse_json": 1.0,
1512
+ "exact_match": 0.0,
1513
+ "field_precision": 0.0,
1514
+ "field_recall": 0.0,
1515
+ "field_f1": 0.0,
1516
+ "field_tp": 0.0,
1517
+ "field_fp": 3.0,
1518
+ "field_fn": 7.0,
1519
+ "slice_sst_pass": 0.0,
1520
+ "kpi_text_presence_pass": 0.0,
1521
+ "adversarial_status_pass": 1.0,
1522
+ "norm_parse_json": 1.0,
1523
+ "norm_gold_parse_json": 1.0,
1524
+ "norm_exact_match": 0.0,
1525
+ "norm_field_precision": 0.0,
1526
+ "norm_field_recall": 0.0,
1527
+ "norm_field_f1": 0.0,
1528
+ "norm_field_tp": 0.0,
1529
+ "norm_field_fp": 3.0,
1530
+ "norm_field_fn": 6.0,
1531
+ "norm_key_precision": 0.0,
1532
+ "norm_key_recall": 0.0,
1533
+ "norm_key_f1": 0.0,
1534
+ "norm_key_tp": 0.0,
1535
+ "norm_key_fp": 3.0,
1536
+ "norm_key_fn": 6.0
1537
+ }
1538
+ },
1539
+ "by_slice_type": {
1540
+ "HMTC": {
1541
+ "num_examples": 28,
1542
+ "parse_json": 0.2857142857142857,
1543
+ "gold_parse_json": 1.0,
1544
+ "exact_match": 0.0,
1545
+ "field_precision": 0.0,
1546
+ "field_recall": 0.0,
1547
+ "field_f1": 0.0,
1548
+ "field_tp": 0.0,
1549
+ "field_fp": 5.785714285714286,
1550
+ "field_fn": 8.892857142857142,
1551
+ "slice_sst_pass": 0.2857142857142857,
1552
+ "kpi_text_presence_pass": 0.25,
1553
+ "adversarial_status_pass": 0.2857142857142857,
1554
+ "norm_parse_json": 0.2857142857142857,
1555
+ "norm_gold_parse_json": 1.0,
1556
+ "norm_exact_match": 0.0,
1557
+ "norm_field_precision": 0.0,
1558
+ "norm_field_recall": 0.0,
1559
+ "norm_field_f1": 0.0,
1560
+ "norm_key_precision": 0.003727753727753728,
1561
+ "norm_key_recall": 0.0019902697921273774,
1562
+ "norm_key_f1": 0.0025097956158000633
1563
+ },
1564
+ "MPS": {
1565
+ "num_examples": 23,
1566
+ "parse_json": 0.2608695652173913,
1567
+ "gold_parse_json": 1.0,
1568
+ "exact_match": 0.0,
1569
+ "field_precision": 0.0,
1570
+ "field_recall": 0.0,
1571
+ "field_f1": 0.0,
1572
+ "field_tp": 0.0,
1573
+ "field_fp": 5.173913043478261,
1574
+ "field_fn": 7.304347826086956,
1575
+ "slice_sst_pass": 0.21739130434782608,
1576
+ "kpi_text_presence_pass": 0.21739130434782608,
1577
+ "adversarial_status_pass": 0.2608695652173913,
1578
+ "norm_parse_json": 0.2608695652173913,
1579
+ "norm_gold_parse_json": 1.0,
1580
+ "norm_exact_match": 0.0,
1581
+ "norm_field_precision": 0.0012422360248447205,
1582
+ "norm_field_recall": 0.0011441647597254005,
1583
+ "norm_field_f1": 0.0011911852293031567,
1584
+ "norm_key_precision": 0.0448334274421231,
1585
+ "norm_key_recall": 0.032494279176201374,
1586
+ "norm_key_f1": 0.03708044342830794
1587
+ },
1588
+ "URLLC": {
1589
+ "num_examples": 30,
1590
+ "parse_json": 0.4,
1591
+ "gold_parse_json": 1.0,
1592
+ "exact_match": 0.0,
1593
+ "field_precision": 0.0,
1594
+ "field_recall": 0.0,
1595
+ "field_f1": 0.0,
1596
+ "field_tp": 0.0,
1597
+ "field_fp": 9.7,
1598
+ "field_fn": 15.133333333333333,
1599
+ "slice_sst_pass": 0.36666666666666664,
1600
+ "kpi_text_presence_pass": 0.36666666666666664,
1601
+ "adversarial_status_pass": 0.4,
1602
+ "norm_parse_json": 0.4,
1603
+ "norm_gold_parse_json": 1.0,
1604
+ "norm_exact_match": 0.0,
1605
+ "norm_field_precision": 0.0,
1606
+ "norm_field_recall": 0.0,
1607
+ "norm_field_f1": 0.0,
1608
+ "norm_key_precision": 0.004693829907029758,
1609
+ "norm_key_recall": 0.002734778121775026,
1610
+ "norm_key_f1": 0.003364717222430942
1611
+ },
1612
+ "V2X": {
1613
+ "num_examples": 42,
1614
+ "parse_json": 0.40476190476190477,
1615
+ "gold_parse_json": 1.0,
1616
+ "exact_match": 0.0,
1617
+ "field_precision": 0.0014217808335455395,
1618
+ "field_recall": 0.0008966244725738396,
1619
+ "field_f1": 0.0010686707115278543,
1620
+ "field_tp": 0.047619047619047616,
1621
+ "field_fp": 11.047619047619047,
1622
+ "field_fn": 15.928571428571429,
1623
+ "slice_sst_pass": 0.07142857142857142,
1624
+ "kpi_text_presence_pass": 0.40476190476190477,
1625
+ "adversarial_status_pass": 0.40476190476190477,
1626
+ "norm_parse_json": 0.40476190476190477,
1627
+ "norm_gold_parse_json": 1.0,
1628
+ "norm_exact_match": 0.0,
1629
+ "norm_field_precision": 0.00745561382447716,
1630
+ "norm_field_recall": 0.005362671384343211,
1631
+ "norm_field_f1": 0.006078850906437114,
1632
+ "norm_key_precision": 0.042990218907293404,
1633
+ "norm_key_recall": 0.03272888102609465,
1634
+ "norm_key_f1": 0.036572723496803475
1635
+ },
1636
+ "eMBB": {
1637
+ "num_examples": 25,
1638
+ "parse_json": 0.4,
1639
+ "gold_parse_json": 1.0,
1640
+ "exact_match": 0.0,
1641
+ "field_precision": 0.0,
1642
+ "field_recall": 0.0,
1643
+ "field_f1": 0.0,
1644
+ "field_tp": 0.0,
1645
+ "field_fp": 6.88,
1646
+ "field_fn": 12.6,
1647
+ "slice_sst_pass": 0.36,
1648
+ "kpi_text_presence_pass": 0.32,
1649
+ "adversarial_status_pass": 0.4,
1650
+ "norm_parse_json": 0.4,
1651
+ "norm_gold_parse_json": 1.0,
1652
+ "norm_exact_match": 0.0,
1653
+ "norm_field_precision": 0.0,
1654
+ "norm_field_recall": 0.0,
1655
+ "norm_field_f1": 0.0,
1656
+ "norm_field_tp": 0.0,
1657
+ "norm_field_fp": 17.2,
1658
+ "norm_field_fn": 29.5,
1659
+ "norm_key_precision": 0.0,
1660
+ "norm_key_recall": 0.0,
1661
+ "norm_key_f1": 0.0,
1662
+ "norm_key_tp": 0.0,
1663
+ "norm_key_fp": 17.2,
1664
+ "norm_key_fn": 29.5
1665
+ },
1666
+ "mMTC": {
1667
+ "num_examples": 52,
1668
+ "parse_json": 0.28846153846153844,
1669
+ "gold_parse_json": 1.0,
1670
+ "exact_match": 0.0,
1671
+ "field_precision": 0.0030048076923076925,
1672
+ "field_recall": 0.002403846153846154,
1673
+ "field_f1": 0.002670940170940171,
1674
+ "field_tp": 0.09615384615384616,
1675
+ "field_fp": 4.25,
1676
+ "field_fn": 6.961538461538462,
1677
+ "slice_sst_pass": 0.11538461538461539,
1678
+ "kpi_text_presence_pass": 0.23076923076923078,
1679
+ "adversarial_status_pass": 0.28846153846153844,
1680
+ "norm_parse_json": 0.28846153846153844,
1681
+ "norm_gold_parse_json": 1.0,
1682
+ "norm_exact_match": 0.0,
1683
+ "norm_field_precision": 0.0,
1684
+ "norm_field_recall": 0.0,
1685
+ "norm_field_f1": 0.0,
1686
+ "norm_field_tp": 0.0,
1687
+ "norm_field_fp": 14.8,
1688
+ "norm_field_fn": 22.666666666666668,
1689
+ "norm_key_precision": 0.020585664335664333,
1690
+ "norm_key_recall": 0.01437246963562753,
1691
+ "norm_key_f1": 0.016728474172642907,
1692
+ "norm_key_tp": 1.5333333333333334,
1693
+ "norm_key_fp": 13.266666666666667,
1694
+ "norm_key_fn": 21.133333333333333
1695
+ }
1696
+ },
1697
+ "by_lifecycle_operation": {
1698
+ "create": {
1699
+ "num_examples": 196,
1700
+ "parse_json": 0.33163265306122447,
1701
+ "gold_parse_json": 1.0,
1702
+ "exact_match": 0.0,
1703
+ "field_precision": 0.0011018611990250646,
1704
+ "field_recall": 0.0008298889175923534,
1705
+ "field_f1": 0.0009376176468013203,
1706
+ "field_tp": 0.03571428571428571,
1707
+ "field_fp": 7.23469387755102,
1708
+ "field_fn": 11.168367346938776,
1709
+ "slice_sst_pass": 0.20408163265306123,
1710
+ "kpi_text_presence_pass": 0.30612244897959184,
1711
+ "adversarial_status_pass": 0.33163265306122447,
1712
+ "norm_parse_json": 0.33163265306122447,
1713
+ "norm_gold_parse_json": 1.0,
1714
+ "norm_exact_match": 0.0,
1715
+ "norm_field_precision": 0.0017434041285687206,
1716
+ "norm_field_recall": 0.0012834081000821379,
1717
+ "norm_field_f1": 0.0014423928486955683,
1718
+ "norm_key_precision": 0.0211857376138658,
1719
+ "norm_key_recall": 0.015342452770581917,
1720
+ "norm_key_f1": 0.017500005288824436
1721
+ },
1722
+ "resume": {
1723
+ "num_examples": 2,
1724
+ "parse_json": 0.5,
1725
+ "gold_parse_json": 1.0,
1726
+ "exact_match": 0.0,
1727
+ "field_precision": 0.0,
1728
+ "field_recall": 0.0,
1729
+ "field_f1": 0.0,
1730
+ "field_tp": 0.0,
1731
+ "field_fp": 1.5,
1732
+ "field_fn": 3.0,
1733
+ "slice_sst_pass": 0.5,
1734
+ "kpi_text_presence_pass": 0.0,
1735
+ "adversarial_status_pass": 0.5,
1736
+ "norm_parse_json": 0.5,
1737
+ "norm_gold_parse_json": 1.0,
1738
+ "norm_exact_match": 0.0,
1739
+ "norm_field_precision": 0.0,
1740
+ "norm_field_recall": 0.0,
1741
+ "norm_field_f1": 0.0,
1742
+ "norm_key_precision": 0.0,
1743
+ "norm_key_recall": 0.0,
1744
+ "norm_key_f1": 0.0
1745
+ },
1746
+ "scale": {
1747
+ "num_examples": 1,
1748
+ "parse_json": 1.0,
1749
+ "gold_parse_json": 1.0,
1750
+ "exact_match": 0.0,
1751
+ "field_precision": 0.0,
1752
+ "field_recall": 0.0,
1753
+ "field_f1": 0.0,
1754
+ "field_tp": 0.0,
1755
+ "field_fp": 5.0,
1756
+ "field_fn": 15.0,
1757
+ "slice_sst_pass": 1.0,
1758
+ "kpi_text_presence_pass": 0.0,
1759
+ "adversarial_status_pass": 1.0,
1760
+ "norm_parse_json": 1.0,
1761
+ "norm_gold_parse_json": 1.0,
1762
+ "norm_exact_match": 0.0,
1763
+ "norm_field_precision": 0.0,
1764
+ "norm_field_recall": 0.0,
1765
+ "norm_field_f1": 0.0,
1766
+ "norm_field_tp": 0.0,
1767
+ "norm_field_fp": 5.0,
1768
+ "norm_field_fn": 13.0,
1769
+ "norm_key_precision": 0.0,
1770
+ "norm_key_recall": 0.0,
1771
+ "norm_key_f1": 0.0,
1772
+ "norm_key_tp": 0.0,
1773
+ "norm_key_fp": 5.0,
1774
+ "norm_key_fn": 13.0
1775
+ },
1776
+ "suspend": {
1777
+ "num_examples": 1,
1778
+ "parse_json": 1.0,
1779
+ "gold_parse_json": 1.0,
1780
+ "exact_match": 0.0,
1781
+ "field_precision": 0.0,
1782
+ "field_recall": 0.0,
1783
+ "field_f1": 0.0,
1784
+ "field_tp": 0.0,
1785
+ "field_fp": 3.0,
1786
+ "field_fn": 7.0,
1787
+ "slice_sst_pass": 0.0,
1788
+ "kpi_text_presence_pass": 0.0,
1789
+ "adversarial_status_pass": 1.0,
1790
+ "norm_parse_json": 1.0,
1791
+ "norm_gold_parse_json": 1.0,
1792
+ "norm_exact_match": 0.0,
1793
+ "norm_field_precision": 0.0,
1794
+ "norm_field_recall": 0.0,
1795
+ "norm_field_f1": 0.0,
1796
+ "norm_field_tp": 0.0,
1797
+ "norm_field_fp": 3.0,
1798
+ "norm_field_fn": 6.0,
1799
+ "norm_key_precision": 0.0,
1800
+ "norm_key_recall": 0.0,
1801
+ "norm_key_f1": 0.0,
1802
+ "norm_key_tp": 0.0,
1803
+ "norm_key_fp": 3.0,
1804
+ "norm_key_fn": 6.0
1805
+ }
1806
+ }
1807
+ },
1808
+ "test_use_case_ood": {
1809
+ "num_examples": 200,
1810
+ "parse_json": 0.325,
1811
+ "gold_parse_json": 1.0,
1812
+ "exact_match": 0.0,
1813
+ "field_precision": 0.0027029176569617745,
1814
+ "field_recall": 0.002125,
1815
+ "field_f1": 0.002374391617133388,
1816
+ "field_tp": 0.085,
1817
+ "field_fp": 6.695,
1818
+ "field_fn": 11.14,
1819
+ "slice_sst_pass": 0.255,
1820
+ "kpi_text_presence_pass": 0.295,
1821
+ "adversarial_status_pass": 0.325,
1822
+ "norm_parse_json": 0.325,
1823
+ "norm_gold_parse_json": 1.0,
1824
+ "norm_exact_match": 0.0,
1825
+ "norm_field_precision": 0.0013040663040663042,
1826
+ "norm_field_recall": 0.0011842105263157893,
1827
+ "norm_field_f1": 0.001236706955016814,
1828
+ "norm_field_tp": 0.13846153846153847,
1829
+ "norm_field_fp": 19.107692307692307,
1830
+ "norm_field_fn": 31.6,
1831
+ "norm_key_precision": 0.024754590082182316,
1832
+ "norm_key_recall": 0.017356037151702787,
1833
+ "norm_key_f1": 0.019825553694296484,
1834
+ "norm_key_tp": 1.8615384615384616,
1835
+ "norm_key_fp": 17.384615384615383,
1836
+ "norm_key_fn": 29.876923076923077,
1837
+ "by_target_layer": {
1838
+ "a1_policy": {
1839
+ "num_examples": 23,
1840
+ "parse_json": 0.17391304347826086,
1841
+ "gold_parse_json": 1.0,
1842
+ "exact_match": 0.0,
1843
+ "field_precision": 0.0,
1844
+ "field_recall": 0.0,
1845
+ "field_f1": 0.0,
1846
+ "field_tp": 0.0,
1847
+ "field_fp": 2.130434782608696,
1848
+ "field_fn": 3.3043478260869565,
1849
+ "slice_sst_pass": 0.08695652173913043,
1850
+ "kpi_text_presence_pass": 0.043478260869565216,
1851
+ "adversarial_status_pass": 0.17391304347826086,
1852
+ "norm_parse_json": 0.17391304347826086,
1853
+ "norm_gold_parse_json": 1.0,
1854
+ "norm_exact_match": 0.0,
1855
+ "norm_field_precision": 0.0,
1856
+ "norm_field_recall": 0.0,
1857
+ "norm_field_f1": 0.0,
1858
+ "norm_key_precision": 0.0,
1859
+ "norm_key_recall": 0.0,
1860
+ "norm_key_f1": 0.0
1861
+ },
1862
+ "camara": {
1863
+ "num_examples": 37,
1864
+ "parse_json": 0.7027027027027027,
1865
+ "gold_parse_json": 1.0,
1866
+ "exact_match": 0.0,
1867
+ "field_precision": 0.0,
1868
+ "field_recall": 0.0,
1869
+ "field_f1": 0.0,
1870
+ "field_tp": 0.0,
1871
+ "field_fp": 5.702702702702703,
1872
+ "field_fn": 11.243243243243244,
1873
+ "slice_sst_pass": 0.5675675675675675,
1874
+ "kpi_text_presence_pass": 0.7027027027027027,
1875
+ "adversarial_status_pass": 0.7027027027027027,
1876
+ "norm_parse_json": 0.7027027027027027,
1877
+ "norm_gold_parse_json": 1.0,
1878
+ "norm_exact_match": 0.0,
1879
+ "norm_field_precision": 0.0,
1880
+ "norm_field_recall": 0.0,
1881
+ "norm_field_f1": 0.0,
1882
+ "norm_field_tp": 0.0,
1883
+ "norm_field_fp": 8.115384615384615,
1884
+ "norm_field_fn": 14.0,
1885
+ "norm_key_precision": 0.0,
1886
+ "norm_key_recall": 0.0,
1887
+ "norm_key_f1": 0.0,
1888
+ "norm_key_tp": 0.0,
1889
+ "norm_key_fp": 8.115384615384615,
1890
+ "norm_key_fn": 14.0
1891
+ },
1892
+ "etsi_zsm": {
1893
+ "num_examples": 20,
1894
+ "parse_json": 0.6,
1895
+ "gold_parse_json": 1.0,
1896
+ "exact_match": 0.0,
1897
+ "field_precision": 0.0,
1898
+ "field_recall": 0.0,
1899
+ "field_f1": 0.0,
1900
+ "field_tp": 0.0,
1901
+ "field_fp": 20.8,
1902
+ "field_fn": 34.2,
1903
+ "slice_sst_pass": 0.5,
1904
+ "kpi_text_presence_pass": 0.6,
1905
+ "adversarial_status_pass": 0.6,
1906
+ "norm_parse_json": 0.6,
1907
+ "norm_gold_parse_json": 1.0,
1908
+ "norm_exact_match": 0.0,
1909
+ "norm_field_precision": 0.0,
1910
+ "norm_field_recall": 0.0,
1911
+ "norm_field_f1": 0.0,
1912
+ "norm_key_precision": 0.0,
1913
+ "norm_key_recall": 0.0,
1914
+ "norm_key_f1": 0.0
1915
+ },
1916
+ "intent_3gpp": {
1917
+ "num_examples": 32,
1918
+ "parse_json": 0.40625,
1919
+ "gold_parse_json": 1.0,
1920
+ "exact_match": 0.0,
1921
+ "field_precision": 0.01689323535601109,
1922
+ "field_recall": 0.01328125,
1923
+ "field_f1": 0.014839947607083674,
1924
+ "field_tp": 0.53125,
1925
+ "field_fp": 13.15625,
1926
+ "field_fn": 15.71875,
1927
+ "slice_sst_pass": 0.34375,
1928
+ "kpi_text_presence_pass": 0.40625,
1929
+ "adversarial_status_pass": 0.40625,
1930
+ "norm_parse_json": 0.40625,
1931
+ "norm_gold_parse_json": 1.0,
1932
+ "norm_exact_match": 0.0,
1933
+ "norm_field_precision": 0.008150414400414401,
1934
+ "norm_field_recall": 0.007401315789473684,
1935
+ "norm_field_f1": 0.0077294184688550885,
1936
+ "norm_field_tp": 0.6923076923076923,
1937
+ "norm_field_fp": 32.92307692307692,
1938
+ "norm_field_fn": 37.30769230769231,
1939
+ "norm_key_precision": 0.08592603325762718,
1940
+ "norm_key_recall": 0.07483552631578948,
1941
+ "norm_key_f1": 0.07975564962394553,
1942
+ "norm_key_tp": 7.0,
1943
+ "norm_key_fp": 26.615384615384617,
1944
+ "norm_key_fn": 31.0
1945
+ },
1946
+ "o1_nrm": {
1947
+ "num_examples": 30,
1948
+ "parse_json": 0.1,
1949
+ "gold_parse_json": 1.0,
1950
+ "exact_match": 0.0,
1951
+ "field_precision": 0.0,
1952
+ "field_recall": 0.0,
1953
+ "field_f1": 0.0,
1954
+ "field_tp": 0.0,
1955
+ "field_fp": 1.4,
1956
+ "field_fn": 2.3,
1957
+ "slice_sst_pass": 0.06666666666666667,
1958
+ "kpi_text_presence_pass": 0.03333333333333333,
1959
+ "adversarial_status_pass": 0.1,
1960
+ "norm_parse_json": 0.1,
1961
+ "norm_gold_parse_json": 1.0,
1962
+ "norm_exact_match": 0.0,
1963
+ "norm_field_precision": 0.0,
1964
+ "norm_field_recall": 0.0,
1965
+ "norm_field_f1": 0.0,
1966
+ "norm_key_precision": 0.055,
1967
+ "norm_key_recall": 0.029999999999999995,
1968
+ "norm_key_f1": 0.038214285714285715
1969
+ },
1970
+ "tmf921": {
1971
+ "num_examples": 49,
1972
+ "parse_json": 0.12244897959183673,
1973
+ "gold_parse_json": 1.0,
1974
+ "exact_match": 0.0,
1975
+ "field_precision": 0.0,
1976
+ "field_recall": 0.0,
1977
+ "field_f1": 0.0,
1978
+ "field_tp": 0.0,
1979
+ "field_fp": 3.9591836734693877,
1980
+ "field_fn": 9.673469387755102,
1981
+ "slice_sst_pass": 0.10204081632653061,
1982
+ "kpi_text_presence_pass": 0.12244897959183673,
1983
+ "adversarial_status_pass": 0.12244897959183673,
1984
+ "norm_parse_json": 0.12244897959183673,
1985
+ "norm_gold_parse_json": 1.0,
1986
+ "norm_exact_match": 0.0,
1987
+ "norm_field_precision": 0.0,
1988
+ "norm_field_recall": 0.0,
1989
+ "norm_field_f1": 0.0,
1990
+ "norm_key_precision": 0.011250713310048841,
1991
+ "norm_key_recall": 0.0036014405762304917,
1992
+ "norm_key_f1": 0.005438803662540167
1993
+ },
1994
+ "tmf921_lifecycle_activate": {
1995
+ "num_examples": 1,
1996
+ "parse_json": 0.0,
1997
+ "gold_parse_json": 1.0,
1998
+ "exact_match": 0.0,
1999
+ "field_precision": 0.0,
2000
+ "field_recall": 0.0,
2001
+ "field_f1": 0.0,
2002
+ "field_tp": 0.0,
2003
+ "field_fp": 0.0,
2004
+ "field_fn": 0.0,
2005
+ "slice_sst_pass": 0.0,
2006
+ "kpi_text_presence_pass": 0.0,
2007
+ "adversarial_status_pass": 0.0,
2008
+ "norm_parse_json": 0.0,
2009
+ "norm_gold_parse_json": 1.0,
2010
+ "norm_exact_match": 0.0,
2011
+ "norm_field_precision": 0.0,
2012
+ "norm_field_recall": 0.0,
2013
+ "norm_field_f1": 0.0,
2014
+ "norm_key_precision": 0.0,
2015
+ "norm_key_recall": 0.0,
2016
+ "norm_key_f1": 0.0
2017
+ },
2018
+ "tmf921_lifecycle_modify": {
2019
+ "num_examples": 3,
2020
+ "parse_json": 0.0,
2021
+ "gold_parse_json": 1.0,
2022
+ "exact_match": 0.0,
2023
+ "field_precision": 0.0,
2024
+ "field_recall": 0.0,
2025
+ "field_f1": 0.0,
2026
+ "field_tp": 0.0,
2027
+ "field_fp": 0.0,
2028
+ "field_fn": 0.0,
2029
+ "slice_sst_pass": 0.0,
2030
+ "kpi_text_presence_pass": 0.0,
2031
+ "adversarial_status_pass": 0.0,
2032
+ "norm_parse_json": 0.0,
2033
+ "norm_gold_parse_json": 1.0,
2034
+ "norm_exact_match": 0.0,
2035
+ "norm_field_precision": 0.0,
2036
+ "norm_field_recall": 0.0,
2037
+ "norm_field_f1": 0.0,
2038
+ "norm_key_precision": 0.0,
2039
+ "norm_key_recall": 0.0,
2040
+ "norm_key_f1": 0.0
2041
+ },
2042
+ "tmf921_lifecycle_resume": {
2043
+ "num_examples": 4,
2044
+ "parse_json": 0.25,
2045
+ "gold_parse_json": 1.0,
2046
+ "exact_match": 0.0,
2047
+ "field_precision": 0.0,
2048
+ "field_recall": 0.0,
2049
+ "field_f1": 0.0,
2050
+ "field_tp": 0.0,
2051
+ "field_fp": 1.5,
2052
+ "field_fn": 1.5,
2053
+ "slice_sst_pass": 0.0,
2054
+ "kpi_text_presence_pass": 0.0,
2055
+ "adversarial_status_pass": 0.25,
2056
+ "norm_parse_json": 0.25,
2057
+ "norm_gold_parse_json": 1.0,
2058
+ "norm_exact_match": 0.0,
2059
+ "norm_field_precision": 0.0,
2060
+ "norm_field_recall": 0.0,
2061
+ "norm_field_f1": 0.0,
2062
+ "norm_key_precision": 0.0,
2063
+ "norm_key_recall": 0.0,
2064
+ "norm_key_f1": 0.0
2065
+ },
2066
+ "tmf921_lifecycle_suspend": {
2067
+ "num_examples": 1,
2068
+ "parse_json": 0.0,
2069
+ "gold_parse_json": 1.0,
2070
+ "exact_match": 0.0,
2071
+ "field_precision": 0.0,
2072
+ "field_recall": 0.0,
2073
+ "field_f1": 0.0,
2074
+ "field_tp": 0.0,
2075
+ "field_fp": 0.0,
2076
+ "field_fn": 0.0,
2077
+ "slice_sst_pass": 0.0,
2078
+ "kpi_text_presence_pass": 0.0,
2079
+ "adversarial_status_pass": 0.0,
2080
+ "norm_parse_json": 0.0,
2081
+ "norm_gold_parse_json": 1.0,
2082
+ "norm_exact_match": 0.0,
2083
+ "norm_field_precision": 0.0,
2084
+ "norm_field_recall": 0.0,
2085
+ "norm_field_f1": 0.0,
2086
+ "norm_key_precision": 0.0,
2087
+ "norm_key_recall": 0.0,
2088
+ "norm_key_f1": 0.0
2089
+ }
2090
+ },
2091
+ "by_slice_type": {
2092
+ "HMTC": {
2093
+ "num_examples": 13,
2094
+ "parse_json": 0.15384615384615385,
2095
+ "gold_parse_json": 1.0,
2096
+ "exact_match": 0.0,
2097
+ "field_precision": 0.0,
2098
+ "field_recall": 0.0,
2099
+ "field_f1": 0.0,
2100
+ "field_tp": 0.0,
2101
+ "field_fp": 2.8461538461538463,
2102
+ "field_fn": 5.615384615384615,
2103
+ "slice_sst_pass": 0.15384615384615385,
2104
+ "kpi_text_presence_pass": 0.15384615384615385,
2105
+ "adversarial_status_pass": 0.15384615384615385,
2106
+ "norm_parse_json": 0.15384615384615385,
2107
+ "norm_gold_parse_json": 1.0,
2108
+ "norm_exact_match": 0.0,
2109
+ "norm_field_precision": 0.0,
2110
+ "norm_field_recall": 0.0,
2111
+ "norm_field_f1": 0.0,
2112
+ "norm_key_precision": 0.0,
2113
+ "norm_key_recall": 0.0,
2114
+ "norm_key_f1": 0.0
2115
+ },
2116
+ "MPS": {
2117
+ "num_examples": 22,
2118
+ "parse_json": 0.5909090909090909,
2119
+ "gold_parse_json": 1.0,
2120
+ "exact_match": 0.0,
2121
+ "field_precision": 0.0,
2122
+ "field_recall": 0.0,
2123
+ "field_f1": 0.0,
2124
+ "field_tp": 0.0,
2125
+ "field_fp": 15.454545454545455,
2126
+ "field_fn": 22.181818181818183,
2127
+ "slice_sst_pass": 0.5909090909090909,
2128
+ "kpi_text_presence_pass": 0.5454545454545454,
2129
+ "adversarial_status_pass": 0.5909090909090909,
2130
+ "norm_parse_json": 0.5909090909090909,
2131
+ "norm_gold_parse_json": 1.0,
2132
+ "norm_exact_match": 0.0,
2133
+ "norm_field_precision": 0.0,
2134
+ "norm_field_recall": 0.0,
2135
+ "norm_field_f1": 0.0,
2136
+ "norm_field_tp": 0.0,
2137
+ "norm_field_fp": 24.846153846153847,
2138
+ "norm_field_fn": 35.38461538461539,
2139
+ "norm_key_precision": 0.04220779220779221,
2140
+ "norm_key_recall": 0.0284688995215311,
2141
+ "norm_key_f1": 0.03387520014232343,
2142
+ "norm_key_tp": 1.0,
2143
+ "norm_key_fp": 23.846153846153847,
2144
+ "norm_key_fn": 34.38461538461539
2145
+ },
2146
+ "URLLC": {
2147
+ "num_examples": 51,
2148
+ "parse_json": 0.35294117647058826,
2149
+ "gold_parse_json": 1.0,
2150
+ "exact_match": 0.0,
2151
+ "field_precision": 0.0,
2152
+ "field_recall": 0.0,
2153
+ "field_f1": 0.0,
2154
+ "field_tp": 0.0,
2155
+ "field_fp": 5.921568627450981,
2156
+ "field_fn": 10.176470588235293,
2157
+ "slice_sst_pass": 0.27450980392156865,
2158
+ "kpi_text_presence_pass": 0.27450980392156865,
2159
+ "adversarial_status_pass": 0.35294117647058826,
2160
+ "norm_parse_json": 0.35294117647058826,
2161
+ "norm_gold_parse_json": 1.0,
2162
+ "norm_exact_match": 0.0,
2163
+ "norm_field_precision": 0.002011060834590246,
2164
+ "norm_field_recall": 0.0020639834881320948,
2165
+ "norm_field_f1": 0.0020371785077667433,
2166
+ "norm_field_tp": 0.2222222222222222,
2167
+ "norm_field_fp": 15.38888888888889,
2168
+ "norm_field_fn": 25.77777777777778,
2169
+ "norm_key_precision": 0.027478099993312976,
2170
+ "norm_key_recall": 0.016839677047289503,
2171
+ "norm_key_f1": 0.01976978148440594,
2172
+ "norm_key_tp": 1.6111111111111112,
2173
+ "norm_key_fp": 14.0,
2174
+ "norm_key_fn": 24.38888888888889
2175
+ },
2176
+ "V2X": {
2177
+ "num_examples": 29,
2178
+ "parse_json": 0.3448275862068966,
2179
+ "gold_parse_json": 1.0,
2180
+ "exact_match": 0.0,
2181
+ "field_precision": 0.0010449320794148381,
2182
+ "field_recall": 0.0008620689655172415,
2183
+ "field_f1": 0.000944733112895607,
2184
+ "field_tp": 0.034482758620689655,
2185
+ "field_fp": 7.275862068965517,
2186
+ "field_fn": 14.0,
2187
+ "slice_sst_pass": 0.2413793103448276,
2188
+ "kpi_text_presence_pass": 0.3448275862068966,
2189
+ "adversarial_status_pass": 0.3448275862068966,
2190
+ "norm_parse_json": 0.3448275862068966,
2191
+ "norm_gold_parse_json": 1.0,
2192
+ "norm_exact_match": 0.0,
2193
+ "norm_field_precision": 0.0041797283176593526,
2194
+ "norm_field_recall": 0.003629764065335753,
2195
+ "norm_field_f1": 0.0038853812530354544,
2196
+ "norm_key_precision": 0.026931749436286644,
2197
+ "norm_key_recall": 0.017561652610227393,
2198
+ "norm_key_f1": 0.0202269043705295
2199
+ },
2200
+ "eMBB": {
2201
+ "num_examples": 64,
2202
+ "parse_json": 0.234375,
2203
+ "gold_parse_json": 1.0,
2204
+ "exact_match": 0.0,
2205
+ "field_precision": 0.005531726579520697,
2206
+ "field_recall": 0.004296875,
2207
+ "field_f1": 0.004821752722872125,
2208
+ "field_tp": 0.171875,
2209
+ "field_fp": 4.671875,
2210
+ "field_fn": 8.046875,
2211
+ "slice_sst_pass": 0.203125,
2212
+ "kpi_text_presence_pass": 0.234375,
2213
+ "adversarial_status_pass": 0.234375,
2214
+ "norm_parse_json": 0.234375,
2215
+ "norm_gold_parse_json": 1.0,
2216
+ "norm_exact_match": 0.0,
2217
+ "norm_field_precision": 0.0005787037037037037,
2218
+ "norm_field_recall": 0.00041118421052631577,
2219
+ "norm_field_f1": 0.0004807692307692308,
2220
+ "norm_key_precision": 0.02047704671637309,
2221
+ "norm_key_recall": 0.01608455882352941,
2222
+ "norm_key_f1": 0.017814009661835748
2223
+ },
2224
+ "mMTC": {
2225
+ "num_examples": 21,
2226
+ "parse_json": 0.3333333333333333,
2227
+ "gold_parse_json": 1.0,
2228
+ "exact_match": 0.0,
2229
+ "field_precision": 0.00744047619047619,
2230
+ "field_recall": 0.005952380952380952,
2231
+ "field_f1": 0.006613756613756614,
2232
+ "field_tp": 0.23809523809523808,
2233
+ "field_fp": 7.142857142857143,
2234
+ "field_fn": 10.80952380952381,
2235
+ "slice_sst_pass": 0.09523809523809523,
2236
+ "kpi_text_presence_pass": 0.2857142857142857,
2237
+ "adversarial_status_pass": 0.3333333333333333,
2238
+ "norm_parse_json": 0.3333333333333333,
2239
+ "norm_gold_parse_json": 1.0,
2240
+ "norm_exact_match": 0.0,
2241
+ "norm_field_precision": 0.0,
2242
+ "norm_field_recall": 0.0,
2243
+ "norm_field_f1": 0.0,
2244
+ "norm_key_precision": 0.025210084033613446,
2245
+ "norm_key_recall": 0.021303258145363407,
2246
+ "norm_key_f1": 0.02309145880574452
2247
+ }
2248
+ },
2249
+ "by_lifecycle_operation": {
2250
+ "activate": {
2251
+ "num_examples": 1,
2252
+ "parse_json": 0.0,
2253
+ "gold_parse_json": 1.0,
2254
+ "exact_match": 0.0,
2255
+ "field_precision": 0.0,
2256
+ "field_recall": 0.0,
2257
+ "field_f1": 0.0,
2258
+ "field_tp": 0.0,
2259
+ "field_fp": 0.0,
2260
+ "field_fn": 0.0,
2261
+ "slice_sst_pass": 0.0,
2262
+ "kpi_text_presence_pass": 0.0,
2263
+ "adversarial_status_pass": 0.0,
2264
+ "norm_parse_json": 0.0,
2265
+ "norm_gold_parse_json": 1.0,
2266
+ "norm_exact_match": 0.0,
2267
+ "norm_field_precision": 0.0,
2268
+ "norm_field_recall": 0.0,
2269
+ "norm_field_f1": 0.0,
2270
+ "norm_key_precision": 0.0,
2271
+ "norm_key_recall": 0.0,
2272
+ "norm_key_f1": 0.0
2273
+ },
2274
+ "create": {
2275
+ "num_examples": 191,
2276
+ "parse_json": 0.33507853403141363,
2277
+ "gold_parse_json": 1.0,
2278
+ "exact_match": 0.0,
2279
+ "field_precision": 0.002830280269069921,
2280
+ "field_recall": 0.002225130890052356,
2281
+ "field_f1": 0.002486273944642291,
2282
+ "field_tp": 0.08900523560209424,
2283
+ "field_fp": 6.979057591623037,
2284
+ "field_fn": 11.633507853403142,
2285
+ "slice_sst_pass": 0.2670157068062827,
2286
+ "kpi_text_presence_pass": 0.3089005235602094,
2287
+ "adversarial_status_pass": 0.33507853403141363,
2288
+ "norm_parse_json": 0.33507853403141363,
2289
+ "norm_gold_parse_json": 1.0,
2290
+ "norm_exact_match": 0.0,
2291
+ "norm_field_precision": 0.0013655144545196903,
2292
+ "norm_field_recall": 0.0012400110223201983,
2293
+ "norm_field_f1": 0.0012949811047296483,
2294
+ "norm_field_tp": 0.140625,
2295
+ "norm_field_fp": 19.34375,
2296
+ "norm_field_fn": 32.015625,
2297
+ "norm_key_precision": 0.02592103673526944,
2298
+ "norm_key_recall": 0.018173860891835376,
2299
+ "norm_key_f1": 0.020759742088268567,
2300
+ "norm_key_tp": 1.890625,
2301
+ "norm_key_fp": 17.59375,
2302
+ "norm_key_fn": 30.265625
2303
+ },
2304
+ "modify": {
2305
+ "num_examples": 3,
2306
+ "parse_json": 0.0,
2307
+ "gold_parse_json": 1.0,
2308
+ "exact_match": 0.0,
2309
+ "field_precision": 0.0,
2310
+ "field_recall": 0.0,
2311
+ "field_f1": 0.0,
2312
+ "field_tp": 0.0,
2313
+ "field_fp": 0.0,
2314
+ "field_fn": 0.0,
2315
+ "slice_sst_pass": 0.0,
2316
+ "kpi_text_presence_pass": 0.0,
2317
+ "adversarial_status_pass": 0.0,
2318
+ "norm_parse_json": 0.0,
2319
+ "norm_gold_parse_json": 1.0,
2320
+ "norm_exact_match": 0.0,
2321
+ "norm_field_precision": 0.0,
2322
+ "norm_field_recall": 0.0,
2323
+ "norm_field_f1": 0.0,
2324
+ "norm_key_precision": 0.0,
2325
+ "norm_key_recall": 0.0,
2326
+ "norm_key_f1": 0.0
2327
+ },
2328
+ "resume": {
2329
+ "num_examples": 4,
2330
+ "parse_json": 0.25,
2331
+ "gold_parse_json": 1.0,
2332
+ "exact_match": 0.0,
2333
+ "field_precision": 0.0,
2334
+ "field_recall": 0.0,
2335
+ "field_f1": 0.0,
2336
+ "field_tp": 0.0,
2337
+ "field_fp": 1.5,
2338
+ "field_fn": 1.5,
2339
+ "slice_sst_pass": 0.0,
2340
+ "kpi_text_presence_pass": 0.0,
2341
+ "adversarial_status_pass": 0.25,
2342
+ "norm_parse_json": 0.25,
2343
+ "norm_gold_parse_json": 1.0,
2344
+ "norm_exact_match": 0.0,
2345
+ "norm_field_precision": 0.0,
2346
+ "norm_field_recall": 0.0,
2347
+ "norm_field_f1": 0.0,
2348
+ "norm_key_precision": 0.0,
2349
+ "norm_key_recall": 0.0,
2350
+ "norm_key_f1": 0.0
2351
+ },
2352
+ "suspend": {
2353
+ "num_examples": 1,
2354
+ "parse_json": 0.0,
2355
+ "gold_parse_json": 1.0,
2356
+ "exact_match": 0.0,
2357
+ "field_precision": 0.0,
2358
+ "field_recall": 0.0,
2359
+ "field_f1": 0.0,
2360
+ "field_tp": 0.0,
2361
+ "field_fp": 0.0,
2362
+ "field_fn": 0.0,
2363
+ "slice_sst_pass": 0.0,
2364
+ "kpi_text_presence_pass": 0.0,
2365
+ "adversarial_status_pass": 0.0,
2366
+ "norm_parse_json": 0.0,
2367
+ "norm_gold_parse_json": 1.0,
2368
+ "norm_exact_match": 0.0,
2369
+ "norm_field_precision": 0.0,
2370
+ "norm_field_recall": 0.0,
2371
+ "norm_field_f1": 0.0,
2372
+ "norm_key_precision": 0.0,
2373
+ "norm_key_recall": 0.0,
2374
+ "norm_key_f1": 0.0
2375
+ }
2376
+ }
2377
+ }
2378
+ }
results/baselines/qwen3_8b_zero_shot_raw_200.json ADDED
@@ -0,0 +1,1412 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "test_in_distribution": {
3
+ "num_examples": 200,
4
+ "parse_json": 0.335,
5
+ "gold_parse_json": 1.0,
6
+ "exact_match": 0.0,
7
+ "field_precision": 0.0015219907407407406,
8
+ "field_recall": 0.0008423913043478261,
9
+ "field_f1": 0.0010584844594233136,
10
+ "field_tp": 0.03,
11
+ "field_fp": 5.855,
12
+ "field_fn": 10.515,
13
+ "slice_sst_pass": 0.285,
14
+ "kpi_text_presence_pass": 0.275,
15
+ "adversarial_status_pass": 0.335,
16
+ "by_target_layer": {
17
+ "a1_policy": {
18
+ "num_examples": 27,
19
+ "parse_json": 0.14814814814814814,
20
+ "gold_parse_json": 1.0,
21
+ "exact_match": 0.0,
22
+ "field_precision": 0.0,
23
+ "field_recall": 0.0,
24
+ "field_f1": 0.0,
25
+ "field_tp": 0.0,
26
+ "field_fp": 1.5555555555555556,
27
+ "field_fn": 2.814814814814815,
28
+ "slice_sst_pass": 0.1111111111111111,
29
+ "kpi_text_presence_pass": 0.07407407407407407,
30
+ "adversarial_status_pass": 0.14814814814814814
31
+ },
32
+ "camara": {
33
+ "num_examples": 36,
34
+ "parse_json": 0.7222222222222222,
35
+ "gold_parse_json": 1.0,
36
+ "exact_match": 0.0,
37
+ "field_precision": 0.0,
38
+ "field_recall": 0.0,
39
+ "field_f1": 0.0,
40
+ "field_tp": 0.0,
41
+ "field_fp": 5.777777777777778,
42
+ "field_fn": 11.61111111111111,
43
+ "slice_sst_pass": 0.5833333333333334,
44
+ "kpi_text_presence_pass": 0.7222222222222222,
45
+ "adversarial_status_pass": 0.7222222222222222
46
+ },
47
+ "etsi_zsm": {
48
+ "num_examples": 20,
49
+ "parse_json": 0.25,
50
+ "gold_parse_json": 1.0,
51
+ "exact_match": 0.0,
52
+ "field_precision": 0.0,
53
+ "field_recall": 0.0,
54
+ "field_f1": 0.0,
55
+ "field_tp": 0.0,
56
+ "field_fp": 8.85,
57
+ "field_fn": 14.25,
58
+ "slice_sst_pass": 0.25,
59
+ "kpi_text_presence_pass": 0.25,
60
+ "adversarial_status_pass": 0.25
61
+ },
62
+ "intent_3gpp": {
63
+ "num_examples": 39,
64
+ "parse_json": 0.46153846153846156,
65
+ "gold_parse_json": 1.0,
66
+ "exact_match": 0.0,
67
+ "field_precision": 0.004599952516619183,
68
+ "field_recall": 0.003205128205128205,
69
+ "field_f1": 0.003773865714164222,
70
+ "field_tp": 0.1282051282051282,
71
+ "field_fp": 12.333333333333334,
72
+ "field_fn": 18.333333333333332,
73
+ "slice_sst_pass": 0.358974358974359,
74
+ "kpi_text_presence_pass": 0.38461538461538464,
75
+ "adversarial_status_pass": 0.46153846153846156
76
+ },
77
+ "o1_nrm": {
78
+ "num_examples": 19,
79
+ "parse_json": 0.21052631578947367,
80
+ "gold_parse_json": 1.0,
81
+ "exact_match": 0.0,
82
+ "field_precision": 0.006578947368421052,
83
+ "field_recall": 0.002288329519450801,
84
+ "field_f1": 0.003395585738539898,
85
+ "field_tp": 0.05263157894736842,
86
+ "field_fp": 2.0526315789473686,
87
+ "field_fn": 4.7894736842105265,
88
+ "slice_sst_pass": 0.21052631578947367,
89
+ "kpi_text_presence_pass": 0.05263157894736842,
90
+ "adversarial_status_pass": 0.21052631578947367
91
+ },
92
+ "tmf921": {
93
+ "num_examples": 45,
94
+ "parse_json": 0.13333333333333333,
95
+ "gold_parse_json": 1.0,
96
+ "exact_match": 0.0,
97
+ "field_precision": 0.0,
98
+ "field_recall": 0.0,
99
+ "field_f1": 0.0,
100
+ "field_tp": 0.0,
101
+ "field_fp": 4.5777777777777775,
102
+ "field_fn": 10.533333333333333,
103
+ "slice_sst_pass": 0.13333333333333333,
104
+ "kpi_text_presence_pass": 0.13333333333333333,
105
+ "adversarial_status_pass": 0.13333333333333333
106
+ },
107
+ "tmf921_lifecycle_activate": {
108
+ "num_examples": 2,
109
+ "parse_json": 0.0,
110
+ "gold_parse_json": 1.0,
111
+ "exact_match": 0.0,
112
+ "field_precision": 0.0,
113
+ "field_recall": 0.0,
114
+ "field_f1": 0.0,
115
+ "field_tp": 0.0,
116
+ "field_fp": 0.0,
117
+ "field_fn": 0.0,
118
+ "slice_sst_pass": 0.0,
119
+ "kpi_text_presence_pass": 0.0,
120
+ "adversarial_status_pass": 0.0
121
+ },
122
+ "tmf921_lifecycle_report": {
123
+ "num_examples": 4,
124
+ "parse_json": 0.0,
125
+ "gold_parse_json": 1.0,
126
+ "exact_match": 0.0,
127
+ "field_precision": 0.0,
128
+ "field_recall": 0.0,
129
+ "field_f1": 0.0,
130
+ "field_tp": 0.0,
131
+ "field_fp": 0.0,
132
+ "field_fn": 0.0,
133
+ "slice_sst_pass": 0.0,
134
+ "kpi_text_presence_pass": 0.0,
135
+ "adversarial_status_pass": 0.0
136
+ },
137
+ "tmf921_lifecycle_resume": {
138
+ "num_examples": 2,
139
+ "parse_json": 0.0,
140
+ "gold_parse_json": 1.0,
141
+ "exact_match": 0.0,
142
+ "field_precision": 0.0,
143
+ "field_recall": 0.0,
144
+ "field_f1": 0.0,
145
+ "field_tp": 0.0,
146
+ "field_fp": 0.0,
147
+ "field_fn": 0.0,
148
+ "slice_sst_pass": 0.0,
149
+ "kpi_text_presence_pass": 0.0,
150
+ "adversarial_status_pass": 0.0
151
+ },
152
+ "tmf921_lifecycle_scale": {
153
+ "num_examples": 3,
154
+ "parse_json": 0.6666666666666666,
155
+ "gold_parse_json": 1.0,
156
+ "exact_match": 0.0,
157
+ "field_precision": 0.0,
158
+ "field_recall": 0.0,
159
+ "field_f1": 0.0,
160
+ "field_tp": 0.0,
161
+ "field_fp": 3.6666666666666665,
162
+ "field_fn": 10.0,
163
+ "slice_sst_pass": 0.6666666666666666,
164
+ "kpi_text_presence_pass": 0.0,
165
+ "adversarial_status_pass": 0.6666666666666666
166
+ },
167
+ "tmf921_lifecycle_suspend": {
168
+ "num_examples": 3,
169
+ "parse_json": 0.6666666666666666,
170
+ "gold_parse_json": 1.0,
171
+ "exact_match": 0.0,
172
+ "field_precision": 0.0,
173
+ "field_recall": 0.0,
174
+ "field_f1": 0.0,
175
+ "field_tp": 0.0,
176
+ "field_fp": 2.3333333333333335,
177
+ "field_fn": 4.666666666666667,
178
+ "slice_sst_pass": 0.6666666666666666,
179
+ "kpi_text_presence_pass": 0.0,
180
+ "adversarial_status_pass": 0.6666666666666666
181
+ }
182
+ },
183
+ "by_slice_type": {
184
+ "HMTC": {
185
+ "num_examples": 13,
186
+ "parse_json": 0.23076923076923078,
187
+ "gold_parse_json": 1.0,
188
+ "exact_match": 0.0,
189
+ "field_precision": 0.0,
190
+ "field_recall": 0.0,
191
+ "field_f1": 0.0,
192
+ "field_tp": 0.0,
193
+ "field_fp": 3.8461538461538463,
194
+ "field_fn": 5.461538461538462,
195
+ "slice_sst_pass": 0.23076923076923078,
196
+ "kpi_text_presence_pass": 0.23076923076923078,
197
+ "adversarial_status_pass": 0.23076923076923078
198
+ },
199
+ "MPS": {
200
+ "num_examples": 28,
201
+ "parse_json": 0.2857142857142857,
202
+ "gold_parse_json": 1.0,
203
+ "exact_match": 0.0,
204
+ "field_precision": 0.005291005291005291,
205
+ "field_recall": 0.0035714285714285718,
206
+ "field_f1": 0.004264392324093817,
207
+ "field_tp": 0.14285714285714285,
208
+ "field_fp": 4.964285714285714,
209
+ "field_fn": 9.071428571428571,
210
+ "slice_sst_pass": 0.25,
211
+ "kpi_text_presence_pass": 0.25,
212
+ "adversarial_status_pass": 0.2857142857142857
213
+ },
214
+ "URLLC": {
215
+ "num_examples": 53,
216
+ "parse_json": 0.39622641509433965,
217
+ "gold_parse_json": 1.0,
218
+ "exact_match": 0.0,
219
+ "field_precision": 0.0005896226415094339,
220
+ "field_recall": 0.0004716981132075472,
221
+ "field_f1": 0.0005241090146750524,
222
+ "field_tp": 0.018867924528301886,
223
+ "field_fp": 8.075471698113208,
224
+ "field_fn": 13.09433962264151,
225
+ "slice_sst_pass": 0.33962264150943394,
226
+ "kpi_text_presence_pass": 0.37735849056603776,
227
+ "adversarial_status_pass": 0.39622641509433965
228
+ },
229
+ "V2X": {
230
+ "num_examples": 23,
231
+ "parse_json": 0.30434782608695654,
232
+ "gold_parse_json": 1.0,
233
+ "exact_match": 0.0,
234
+ "field_precision": 0.0,
235
+ "field_recall": 0.0,
236
+ "field_f1": 0.0,
237
+ "field_tp": 0.0,
238
+ "field_fp": 6.391304347826087,
239
+ "field_fn": 10.478260869565217,
240
+ "slice_sst_pass": 0.21739130434782608,
241
+ "kpi_text_presence_pass": 0.2608695652173913,
242
+ "adversarial_status_pass": 0.30434782608695654
243
+ },
244
+ "eMBB": {
245
+ "num_examples": 62,
246
+ "parse_json": 0.3225806451612903,
247
+ "gold_parse_json": 1.0,
248
+ "exact_match": 0.0,
249
+ "field_precision": 0.0020161290322580645,
250
+ "field_recall": 0.0007012622720897616,
251
+ "field_f1": 0.0010405827263267429,
252
+ "field_tp": 0.016129032258064516,
253
+ "field_fp": 4.661290322580645,
254
+ "field_fn": 10.661290322580646,
255
+ "slice_sst_pass": 0.2903225806451613,
256
+ "kpi_text_presence_pass": 0.22580645161290322,
257
+ "adversarial_status_pass": 0.3225806451612903
258
+ },
259
+ "mMTC": {
260
+ "num_examples": 21,
261
+ "parse_json": 0.38095238095238093,
262
+ "gold_parse_json": 1.0,
263
+ "exact_match": 0.0,
264
+ "field_precision": 0.0,
265
+ "field_recall": 0.0,
266
+ "field_f1": 0.0,
267
+ "field_tp": 0.0,
268
+ "field_fp": 5.619047619047619,
269
+ "field_fn": 8.666666666666666,
270
+ "slice_sst_pass": 0.2857142857142857,
271
+ "kpi_text_presence_pass": 0.23809523809523808,
272
+ "adversarial_status_pass": 0.38095238095238093
273
+ }
274
+ },
275
+ "by_lifecycle_operation": {
276
+ "activate": {
277
+ "num_examples": 2,
278
+ "parse_json": 0.0,
279
+ "gold_parse_json": 1.0,
280
+ "exact_match": 0.0,
281
+ "field_precision": 0.0,
282
+ "field_recall": 0.0,
283
+ "field_f1": 0.0,
284
+ "field_tp": 0.0,
285
+ "field_fp": 0.0,
286
+ "field_fn": 0.0,
287
+ "slice_sst_pass": 0.0,
288
+ "kpi_text_presence_pass": 0.0,
289
+ "adversarial_status_pass": 0.0
290
+ },
291
+ "create": {
292
+ "num_examples": 186,
293
+ "parse_json": 0.3387096774193548,
294
+ "gold_parse_json": 1.0,
295
+ "exact_match": 0.0,
296
+ "field_precision": 0.0016365491835921943,
297
+ "field_recall": 0.0009057971014492754,
298
+ "field_f1": 0.0011381553327132405,
299
+ "field_tp": 0.03225806451612903,
300
+ "field_fp": 6.198924731182796,
301
+ "field_fn": 11.06989247311828,
302
+ "slice_sst_pass": 0.2849462365591398,
303
+ "kpi_text_presence_pass": 0.2956989247311828,
304
+ "adversarial_status_pass": 0.3387096774193548
305
+ },
306
+ "report": {
307
+ "num_examples": 4,
308
+ "parse_json": 0.0,
309
+ "gold_parse_json": 1.0,
310
+ "exact_match": 0.0,
311
+ "field_precision": 0.0,
312
+ "field_recall": 0.0,
313
+ "field_f1": 0.0,
314
+ "field_tp": 0.0,
315
+ "field_fp": 0.0,
316
+ "field_fn": 0.0,
317
+ "slice_sst_pass": 0.0,
318
+ "kpi_text_presence_pass": 0.0,
319
+ "adversarial_status_pass": 0.0
320
+ },
321
+ "resume": {
322
+ "num_examples": 2,
323
+ "parse_json": 0.0,
324
+ "gold_parse_json": 1.0,
325
+ "exact_match": 0.0,
326
+ "field_precision": 0.0,
327
+ "field_recall": 0.0,
328
+ "field_f1": 0.0,
329
+ "field_tp": 0.0,
330
+ "field_fp": 0.0,
331
+ "field_fn": 0.0,
332
+ "slice_sst_pass": 0.0,
333
+ "kpi_text_presence_pass": 0.0,
334
+ "adversarial_status_pass": 0.0
335
+ },
336
+ "scale": {
337
+ "num_examples": 3,
338
+ "parse_json": 0.6666666666666666,
339
+ "gold_parse_json": 1.0,
340
+ "exact_match": 0.0,
341
+ "field_precision": 0.0,
342
+ "field_recall": 0.0,
343
+ "field_f1": 0.0,
344
+ "field_tp": 0.0,
345
+ "field_fp": 3.6666666666666665,
346
+ "field_fn": 10.0,
347
+ "slice_sst_pass": 0.6666666666666666,
348
+ "kpi_text_presence_pass": 0.0,
349
+ "adversarial_status_pass": 0.6666666666666666
350
+ },
351
+ "suspend": {
352
+ "num_examples": 3,
353
+ "parse_json": 0.6666666666666666,
354
+ "gold_parse_json": 1.0,
355
+ "exact_match": 0.0,
356
+ "field_precision": 0.0,
357
+ "field_recall": 0.0,
358
+ "field_f1": 0.0,
359
+ "field_tp": 0.0,
360
+ "field_fp": 2.3333333333333335,
361
+ "field_fn": 4.666666666666667,
362
+ "slice_sst_pass": 0.6666666666666666,
363
+ "kpi_text_presence_pass": 0.0,
364
+ "adversarial_status_pass": 0.6666666666666666
365
+ }
366
+ }
367
+ },
368
+ "test_template_ood": {
369
+ "num_examples": 200,
370
+ "parse_json": 0.34,
371
+ "gold_parse_json": 1.0,
372
+ "exact_match": 0.0,
373
+ "field_precision": 0.0010798239750445633,
374
+ "field_recall": 0.0008132911392405063,
375
+ "field_f1": 0.0009188652938652939,
376
+ "field_tp": 0.035,
377
+ "field_fp": 7.145,
378
+ "field_fn": 11.085,
379
+ "slice_sst_pass": 0.21,
380
+ "kpi_text_presence_pass": 0.3,
381
+ "adversarial_status_pass": 0.34,
382
+ "by_target_layer": {
383
+ "a1_policy": {
384
+ "num_examples": 33,
385
+ "parse_json": 0.2727272727272727,
386
+ "gold_parse_json": 1.0,
387
+ "exact_match": 0.0,
388
+ "field_precision": 0.0,
389
+ "field_recall": 0.0,
390
+ "field_f1": 0.0,
391
+ "field_tp": 0.0,
392
+ "field_fp": 3.3333333333333335,
393
+ "field_fn": 5.181818181818182,
394
+ "slice_sst_pass": 0.24242424242424243,
395
+ "kpi_text_presence_pass": 0.21212121212121213,
396
+ "adversarial_status_pass": 0.2727272727272727
397
+ },
398
+ "camara": {
399
+ "num_examples": 29,
400
+ "parse_json": 0.7241379310344828,
401
+ "gold_parse_json": 1.0,
402
+ "exact_match": 0.0,
403
+ "field_precision": 0.0,
404
+ "field_recall": 0.0,
405
+ "field_f1": 0.0,
406
+ "field_tp": 0.0,
407
+ "field_fp": 5.862068965517241,
408
+ "field_fn": 11.482758620689655,
409
+ "slice_sst_pass": 0.41379310344827586,
410
+ "kpi_text_presence_pass": 0.7241379310344828,
411
+ "adversarial_status_pass": 0.7241379310344828
412
+ },
413
+ "etsi_zsm": {
414
+ "num_examples": 21,
415
+ "parse_json": 0.47619047619047616,
416
+ "gold_parse_json": 1.0,
417
+ "exact_match": 0.0,
418
+ "field_precision": 0.0,
419
+ "field_recall": 0.0,
420
+ "field_f1": 0.0,
421
+ "field_tp": 0.0,
422
+ "field_fp": 18.714285714285715,
423
+ "field_fn": 27.142857142857142,
424
+ "slice_sst_pass": 0.3333333333333333,
425
+ "kpi_text_presence_pass": 0.47619047619047616,
426
+ "adversarial_status_pass": 0.47619047619047616
427
+ },
428
+ "intent_3gpp": {
429
+ "num_examples": 50,
430
+ "parse_json": 0.38,
431
+ "gold_parse_json": 1.0,
432
+ "exact_match": 0.0,
433
+ "field_precision": 0.0037132352941176474,
434
+ "field_recall": 0.003,
435
+ "field_f1": 0.0033183183183183185,
436
+ "field_tp": 0.12,
437
+ "field_fp": 11.62,
438
+ "field_fn": 15.08,
439
+ "slice_sst_pass": 0.18,
440
+ "kpi_text_presence_pass": 0.36,
441
+ "adversarial_status_pass": 0.38
442
+ },
443
+ "o1_nrm": {
444
+ "num_examples": 15,
445
+ "parse_json": 0.13333333333333333,
446
+ "gold_parse_json": 1.0,
447
+ "exact_match": 0.0,
448
+ "field_precision": 0.0,
449
+ "field_recall": 0.0,
450
+ "field_f1": 0.0,
451
+ "field_tp": 0.0,
452
+ "field_fp": 2.0,
453
+ "field_fn": 3.066666666666667,
454
+ "slice_sst_pass": 0.06666666666666667,
455
+ "kpi_text_presence_pass": 0.0,
456
+ "adversarial_status_pass": 0.13333333333333333
457
+ },
458
+ "tmf921": {
459
+ "num_examples": 48,
460
+ "parse_json": 0.08333333333333333,
461
+ "gold_parse_json": 1.0,
462
+ "exact_match": 0.0,
463
+ "field_precision": 0.0006313131313131314,
464
+ "field_recall": 0.00026371308016877635,
465
+ "field_f1": 0.0003720238095238095,
466
+ "field_tp": 0.020833333333333332,
467
+ "field_fp": 2.7916666666666665,
468
+ "field_fn": 6.5625,
469
+ "slice_sst_pass": 0.0625,
470
+ "kpi_text_presence_pass": 0.08333333333333333,
471
+ "adversarial_status_pass": 0.08333333333333333
472
+ },
473
+ "tmf921_lifecycle_resume": {
474
+ "num_examples": 2,
475
+ "parse_json": 0.5,
476
+ "gold_parse_json": 1.0,
477
+ "exact_match": 0.0,
478
+ "field_precision": 0.0,
479
+ "field_recall": 0.0,
480
+ "field_f1": 0.0,
481
+ "field_tp": 0.0,
482
+ "field_fp": 1.5,
483
+ "field_fn": 3.0,
484
+ "slice_sst_pass": 0.5,
485
+ "kpi_text_presence_pass": 0.0,
486
+ "adversarial_status_pass": 0.5
487
+ },
488
+ "tmf921_lifecycle_scale": {
489
+ "num_examples": 1,
490
+ "parse_json": 1.0,
491
+ "gold_parse_json": 1.0,
492
+ "exact_match": 0.0,
493
+ "field_precision": 0.0,
494
+ "field_recall": 0.0,
495
+ "field_f1": 0.0,
496
+ "field_tp": 0.0,
497
+ "field_fp": 5.0,
498
+ "field_fn": 15.0,
499
+ "slice_sst_pass": 1.0,
500
+ "kpi_text_presence_pass": 0.0,
501
+ "adversarial_status_pass": 1.0
502
+ },
503
+ "tmf921_lifecycle_suspend": {
504
+ "num_examples": 1,
505
+ "parse_json": 1.0,
506
+ "gold_parse_json": 1.0,
507
+ "exact_match": 0.0,
508
+ "field_precision": 0.0,
509
+ "field_recall": 0.0,
510
+ "field_f1": 0.0,
511
+ "field_tp": 0.0,
512
+ "field_fp": 3.0,
513
+ "field_fn": 7.0,
514
+ "slice_sst_pass": 0.0,
515
+ "kpi_text_presence_pass": 0.0,
516
+ "adversarial_status_pass": 1.0
517
+ }
518
+ },
519
+ "by_slice_type": {
520
+ "HMTC": {
521
+ "num_examples": 28,
522
+ "parse_json": 0.2857142857142857,
523
+ "gold_parse_json": 1.0,
524
+ "exact_match": 0.0,
525
+ "field_precision": 0.0,
526
+ "field_recall": 0.0,
527
+ "field_f1": 0.0,
528
+ "field_tp": 0.0,
529
+ "field_fp": 5.785714285714286,
530
+ "field_fn": 8.892857142857142,
531
+ "slice_sst_pass": 0.2857142857142857,
532
+ "kpi_text_presence_pass": 0.25,
533
+ "adversarial_status_pass": 0.2857142857142857
534
+ },
535
+ "MPS": {
536
+ "num_examples": 23,
537
+ "parse_json": 0.2608695652173913,
538
+ "gold_parse_json": 1.0,
539
+ "exact_match": 0.0,
540
+ "field_precision": 0.0,
541
+ "field_recall": 0.0,
542
+ "field_f1": 0.0,
543
+ "field_tp": 0.0,
544
+ "field_fp": 5.173913043478261,
545
+ "field_fn": 7.304347826086956,
546
+ "slice_sst_pass": 0.21739130434782608,
547
+ "kpi_text_presence_pass": 0.21739130434782608,
548
+ "adversarial_status_pass": 0.2608695652173913
549
+ },
550
+ "URLLC": {
551
+ "num_examples": 30,
552
+ "parse_json": 0.4,
553
+ "gold_parse_json": 1.0,
554
+ "exact_match": 0.0,
555
+ "field_precision": 0.0,
556
+ "field_recall": 0.0,
557
+ "field_f1": 0.0,
558
+ "field_tp": 0.0,
559
+ "field_fp": 9.7,
560
+ "field_fn": 15.133333333333333,
561
+ "slice_sst_pass": 0.36666666666666664,
562
+ "kpi_text_presence_pass": 0.36666666666666664,
563
+ "adversarial_status_pass": 0.4
564
+ },
565
+ "V2X": {
566
+ "num_examples": 42,
567
+ "parse_json": 0.40476190476190477,
568
+ "gold_parse_json": 1.0,
569
+ "exact_match": 0.0,
570
+ "field_precision": 0.0014217808335455395,
571
+ "field_recall": 0.0008966244725738396,
572
+ "field_f1": 0.0010686707115278543,
573
+ "field_tp": 0.047619047619047616,
574
+ "field_fp": 11.047619047619047,
575
+ "field_fn": 15.928571428571429,
576
+ "slice_sst_pass": 0.07142857142857142,
577
+ "kpi_text_presence_pass": 0.40476190476190477,
578
+ "adversarial_status_pass": 0.40476190476190477
579
+ },
580
+ "eMBB": {
581
+ "num_examples": 25,
582
+ "parse_json": 0.4,
583
+ "gold_parse_json": 1.0,
584
+ "exact_match": 0.0,
585
+ "field_precision": 0.0,
586
+ "field_recall": 0.0,
587
+ "field_f1": 0.0,
588
+ "field_tp": 0.0,
589
+ "field_fp": 6.88,
590
+ "field_fn": 12.6,
591
+ "slice_sst_pass": 0.36,
592
+ "kpi_text_presence_pass": 0.32,
593
+ "adversarial_status_pass": 0.4
594
+ },
595
+ "mMTC": {
596
+ "num_examples": 52,
597
+ "parse_json": 0.28846153846153844,
598
+ "gold_parse_json": 1.0,
599
+ "exact_match": 0.0,
600
+ "field_precision": 0.0030048076923076925,
601
+ "field_recall": 0.002403846153846154,
602
+ "field_f1": 0.002670940170940171,
603
+ "field_tp": 0.09615384615384616,
604
+ "field_fp": 4.25,
605
+ "field_fn": 6.961538461538462,
606
+ "slice_sst_pass": 0.11538461538461539,
607
+ "kpi_text_presence_pass": 0.23076923076923078,
608
+ "adversarial_status_pass": 0.28846153846153844
609
+ }
610
+ },
611
+ "by_lifecycle_operation": {
612
+ "create": {
613
+ "num_examples": 196,
614
+ "parse_json": 0.33163265306122447,
615
+ "gold_parse_json": 1.0,
616
+ "exact_match": 0.0,
617
+ "field_precision": 0.0011018611990250646,
618
+ "field_recall": 0.0008298889175923534,
619
+ "field_f1": 0.0009376176468013203,
620
+ "field_tp": 0.03571428571428571,
621
+ "field_fp": 7.23469387755102,
622
+ "field_fn": 11.168367346938776,
623
+ "slice_sst_pass": 0.20408163265306123,
624
+ "kpi_text_presence_pass": 0.30612244897959184,
625
+ "adversarial_status_pass": 0.33163265306122447
626
+ },
627
+ "resume": {
628
+ "num_examples": 2,
629
+ "parse_json": 0.5,
630
+ "gold_parse_json": 1.0,
631
+ "exact_match": 0.0,
632
+ "field_precision": 0.0,
633
+ "field_recall": 0.0,
634
+ "field_f1": 0.0,
635
+ "field_tp": 0.0,
636
+ "field_fp": 1.5,
637
+ "field_fn": 3.0,
638
+ "slice_sst_pass": 0.5,
639
+ "kpi_text_presence_pass": 0.0,
640
+ "adversarial_status_pass": 0.5
641
+ },
642
+ "scale": {
643
+ "num_examples": 1,
644
+ "parse_json": 1.0,
645
+ "gold_parse_json": 1.0,
646
+ "exact_match": 0.0,
647
+ "field_precision": 0.0,
648
+ "field_recall": 0.0,
649
+ "field_f1": 0.0,
650
+ "field_tp": 0.0,
651
+ "field_fp": 5.0,
652
+ "field_fn": 15.0,
653
+ "slice_sst_pass": 1.0,
654
+ "kpi_text_presence_pass": 0.0,
655
+ "adversarial_status_pass": 1.0
656
+ },
657
+ "suspend": {
658
+ "num_examples": 1,
659
+ "parse_json": 1.0,
660
+ "gold_parse_json": 1.0,
661
+ "exact_match": 0.0,
662
+ "field_precision": 0.0,
663
+ "field_recall": 0.0,
664
+ "field_f1": 0.0,
665
+ "field_tp": 0.0,
666
+ "field_fp": 3.0,
667
+ "field_fn": 7.0,
668
+ "slice_sst_pass": 0.0,
669
+ "kpi_text_presence_pass": 0.0,
670
+ "adversarial_status_pass": 1.0
671
+ }
672
+ }
673
+ },
674
+ "test_use_case_ood": {
675
+ "num_examples": 200,
676
+ "parse_json": 0.325,
677
+ "gold_parse_json": 1.0,
678
+ "exact_match": 0.0,
679
+ "field_precision": 0.0027029176569617745,
680
+ "field_recall": 0.002125,
681
+ "field_f1": 0.002374391617133388,
682
+ "field_tp": 0.085,
683
+ "field_fp": 6.695,
684
+ "field_fn": 11.14,
685
+ "slice_sst_pass": 0.255,
686
+ "kpi_text_presence_pass": 0.295,
687
+ "adversarial_status_pass": 0.325,
688
+ "by_target_layer": {
689
+ "a1_policy": {
690
+ "num_examples": 23,
691
+ "parse_json": 0.17391304347826086,
692
+ "gold_parse_json": 1.0,
693
+ "exact_match": 0.0,
694
+ "field_precision": 0.0,
695
+ "field_recall": 0.0,
696
+ "field_f1": 0.0,
697
+ "field_tp": 0.0,
698
+ "field_fp": 2.130434782608696,
699
+ "field_fn": 3.3043478260869565,
700
+ "slice_sst_pass": 0.08695652173913043,
701
+ "kpi_text_presence_pass": 0.043478260869565216,
702
+ "adversarial_status_pass": 0.17391304347826086
703
+ },
704
+ "camara": {
705
+ "num_examples": 37,
706
+ "parse_json": 0.7027027027027027,
707
+ "gold_parse_json": 1.0,
708
+ "exact_match": 0.0,
709
+ "field_precision": 0.0,
710
+ "field_recall": 0.0,
711
+ "field_f1": 0.0,
712
+ "field_tp": 0.0,
713
+ "field_fp": 5.702702702702703,
714
+ "field_fn": 11.243243243243244,
715
+ "slice_sst_pass": 0.5675675675675675,
716
+ "kpi_text_presence_pass": 0.7027027027027027,
717
+ "adversarial_status_pass": 0.7027027027027027
718
+ },
719
+ "etsi_zsm": {
720
+ "num_examples": 20,
721
+ "parse_json": 0.6,
722
+ "gold_parse_json": 1.0,
723
+ "exact_match": 0.0,
724
+ "field_precision": 0.0,
725
+ "field_recall": 0.0,
726
+ "field_f1": 0.0,
727
+ "field_tp": 0.0,
728
+ "field_fp": 20.8,
729
+ "field_fn": 34.2,
730
+ "slice_sst_pass": 0.5,
731
+ "kpi_text_presence_pass": 0.6,
732
+ "adversarial_status_pass": 0.6
733
+ },
734
+ "intent_3gpp": {
735
+ "num_examples": 32,
736
+ "parse_json": 0.40625,
737
+ "gold_parse_json": 1.0,
738
+ "exact_match": 0.0,
739
+ "field_precision": 0.01689323535601109,
740
+ "field_recall": 0.01328125,
741
+ "field_f1": 0.014839947607083674,
742
+ "field_tp": 0.53125,
743
+ "field_fp": 13.15625,
744
+ "field_fn": 15.71875,
745
+ "slice_sst_pass": 0.34375,
746
+ "kpi_text_presence_pass": 0.40625,
747
+ "adversarial_status_pass": 0.40625
748
+ },
749
+ "o1_nrm": {
750
+ "num_examples": 30,
751
+ "parse_json": 0.1,
752
+ "gold_parse_json": 1.0,
753
+ "exact_match": 0.0,
754
+ "field_precision": 0.0,
755
+ "field_recall": 0.0,
756
+ "field_f1": 0.0,
757
+ "field_tp": 0.0,
758
+ "field_fp": 1.4,
759
+ "field_fn": 2.3,
760
+ "slice_sst_pass": 0.06666666666666667,
761
+ "kpi_text_presence_pass": 0.03333333333333333,
762
+ "adversarial_status_pass": 0.1
763
+ },
764
+ "tmf921": {
765
+ "num_examples": 49,
766
+ "parse_json": 0.12244897959183673,
767
+ "gold_parse_json": 1.0,
768
+ "exact_match": 0.0,
769
+ "field_precision": 0.0,
770
+ "field_recall": 0.0,
771
+ "field_f1": 0.0,
772
+ "field_tp": 0.0,
773
+ "field_fp": 3.9591836734693877,
774
+ "field_fn": 9.673469387755102,
775
+ "slice_sst_pass": 0.10204081632653061,
776
+ "kpi_text_presence_pass": 0.12244897959183673,
777
+ "adversarial_status_pass": 0.12244897959183673
778
+ },
779
+ "tmf921_lifecycle_activate": {
780
+ "num_examples": 1,
781
+ "parse_json": 0.0,
782
+ "gold_parse_json": 1.0,
783
+ "exact_match": 0.0,
784
+ "field_precision": 0.0,
785
+ "field_recall": 0.0,
786
+ "field_f1": 0.0,
787
+ "field_tp": 0.0,
788
+ "field_fp": 0.0,
789
+ "field_fn": 0.0,
790
+ "slice_sst_pass": 0.0,
791
+ "kpi_text_presence_pass": 0.0,
792
+ "adversarial_status_pass": 0.0
793
+ },
794
+ "tmf921_lifecycle_modify": {
795
+ "num_examples": 3,
796
+ "parse_json": 0.0,
797
+ "gold_parse_json": 1.0,
798
+ "exact_match": 0.0,
799
+ "field_precision": 0.0,
800
+ "field_recall": 0.0,
801
+ "field_f1": 0.0,
802
+ "field_tp": 0.0,
803
+ "field_fp": 0.0,
804
+ "field_fn": 0.0,
805
+ "slice_sst_pass": 0.0,
806
+ "kpi_text_presence_pass": 0.0,
807
+ "adversarial_status_pass": 0.0
808
+ },
809
+ "tmf921_lifecycle_resume": {
810
+ "num_examples": 4,
811
+ "parse_json": 0.25,
812
+ "gold_parse_json": 1.0,
813
+ "exact_match": 0.0,
814
+ "field_precision": 0.0,
815
+ "field_recall": 0.0,
816
+ "field_f1": 0.0,
817
+ "field_tp": 0.0,
818
+ "field_fp": 1.5,
819
+ "field_fn": 1.5,
820
+ "slice_sst_pass": 0.0,
821
+ "kpi_text_presence_pass": 0.0,
822
+ "adversarial_status_pass": 0.25
823
+ },
824
+ "tmf921_lifecycle_suspend": {
825
+ "num_examples": 1,
826
+ "parse_json": 0.0,
827
+ "gold_parse_json": 1.0,
828
+ "exact_match": 0.0,
829
+ "field_precision": 0.0,
830
+ "field_recall": 0.0,
831
+ "field_f1": 0.0,
832
+ "field_tp": 0.0,
833
+ "field_fp": 0.0,
834
+ "field_fn": 0.0,
835
+ "slice_sst_pass": 0.0,
836
+ "kpi_text_presence_pass": 0.0,
837
+ "adversarial_status_pass": 0.0
838
+ }
839
+ },
840
+ "by_slice_type": {
841
+ "HMTC": {
842
+ "num_examples": 13,
843
+ "parse_json": 0.15384615384615385,
844
+ "gold_parse_json": 1.0,
845
+ "exact_match": 0.0,
846
+ "field_precision": 0.0,
847
+ "field_recall": 0.0,
848
+ "field_f1": 0.0,
849
+ "field_tp": 0.0,
850
+ "field_fp": 2.8461538461538463,
851
+ "field_fn": 5.615384615384615,
852
+ "slice_sst_pass": 0.15384615384615385,
853
+ "kpi_text_presence_pass": 0.15384615384615385,
854
+ "adversarial_status_pass": 0.15384615384615385
855
+ },
856
+ "MPS": {
857
+ "num_examples": 22,
858
+ "parse_json": 0.5909090909090909,
859
+ "gold_parse_json": 1.0,
860
+ "exact_match": 0.0,
861
+ "field_precision": 0.0,
862
+ "field_recall": 0.0,
863
+ "field_f1": 0.0,
864
+ "field_tp": 0.0,
865
+ "field_fp": 15.454545454545455,
866
+ "field_fn": 22.181818181818183,
867
+ "slice_sst_pass": 0.5909090909090909,
868
+ "kpi_text_presence_pass": 0.5454545454545454,
869
+ "adversarial_status_pass": 0.5909090909090909
870
+ },
871
+ "URLLC": {
872
+ "num_examples": 51,
873
+ "parse_json": 0.35294117647058826,
874
+ "gold_parse_json": 1.0,
875
+ "exact_match": 0.0,
876
+ "field_precision": 0.0,
877
+ "field_recall": 0.0,
878
+ "field_f1": 0.0,
879
+ "field_tp": 0.0,
880
+ "field_fp": 5.921568627450981,
881
+ "field_fn": 10.176470588235293,
882
+ "slice_sst_pass": 0.27450980392156865,
883
+ "kpi_text_presence_pass": 0.27450980392156865,
884
+ "adversarial_status_pass": 0.35294117647058826
885
+ },
886
+ "V2X": {
887
+ "num_examples": 29,
888
+ "parse_json": 0.3448275862068966,
889
+ "gold_parse_json": 1.0,
890
+ "exact_match": 0.0,
891
+ "field_precision": 0.0010449320794148381,
892
+ "field_recall": 0.0008620689655172415,
893
+ "field_f1": 0.000944733112895607,
894
+ "field_tp": 0.034482758620689655,
895
+ "field_fp": 7.275862068965517,
896
+ "field_fn": 14.0,
897
+ "slice_sst_pass": 0.2413793103448276,
898
+ "kpi_text_presence_pass": 0.3448275862068966,
899
+ "adversarial_status_pass": 0.3448275862068966
900
+ },
901
+ "eMBB": {
902
+ "num_examples": 64,
903
+ "parse_json": 0.234375,
904
+ "gold_parse_json": 1.0,
905
+ "exact_match": 0.0,
906
+ "field_precision": 0.005531726579520697,
907
+ "field_recall": 0.004296875,
908
+ "field_f1": 0.004821752722872125,
909
+ "field_tp": 0.171875,
910
+ "field_fp": 4.671875,
911
+ "field_fn": 8.046875,
912
+ "slice_sst_pass": 0.203125,
913
+ "kpi_text_presence_pass": 0.234375,
914
+ "adversarial_status_pass": 0.234375
915
+ },
916
+ "mMTC": {
917
+ "num_examples": 21,
918
+ "parse_json": 0.3333333333333333,
919
+ "gold_parse_json": 1.0,
920
+ "exact_match": 0.0,
921
+ "field_precision": 0.00744047619047619,
922
+ "field_recall": 0.005952380952380952,
923
+ "field_f1": 0.006613756613756614,
924
+ "field_tp": 0.23809523809523808,
925
+ "field_fp": 7.142857142857143,
926
+ "field_fn": 10.80952380952381,
927
+ "slice_sst_pass": 0.09523809523809523,
928
+ "kpi_text_presence_pass": 0.2857142857142857,
929
+ "adversarial_status_pass": 0.3333333333333333
930
+ }
931
+ },
932
+ "by_lifecycle_operation": {
933
+ "activate": {
934
+ "num_examples": 1,
935
+ "parse_json": 0.0,
936
+ "gold_parse_json": 1.0,
937
+ "exact_match": 0.0,
938
+ "field_precision": 0.0,
939
+ "field_recall": 0.0,
940
+ "field_f1": 0.0,
941
+ "field_tp": 0.0,
942
+ "field_fp": 0.0,
943
+ "field_fn": 0.0,
944
+ "slice_sst_pass": 0.0,
945
+ "kpi_text_presence_pass": 0.0,
946
+ "adversarial_status_pass": 0.0
947
+ },
948
+ "create": {
949
+ "num_examples": 191,
950
+ "parse_json": 0.33507853403141363,
951
+ "gold_parse_json": 1.0,
952
+ "exact_match": 0.0,
953
+ "field_precision": 0.002830280269069921,
954
+ "field_recall": 0.002225130890052356,
955
+ "field_f1": 0.002486273944642291,
956
+ "field_tp": 0.08900523560209424,
957
+ "field_fp": 6.979057591623037,
958
+ "field_fn": 11.633507853403142,
959
+ "slice_sst_pass": 0.2670157068062827,
960
+ "kpi_text_presence_pass": 0.3089005235602094,
961
+ "adversarial_status_pass": 0.33507853403141363
962
+ },
963
+ "modify": {
964
+ "num_examples": 3,
965
+ "parse_json": 0.0,
966
+ "gold_parse_json": 1.0,
967
+ "exact_match": 0.0,
968
+ "field_precision": 0.0,
969
+ "field_recall": 0.0,
970
+ "field_f1": 0.0,
971
+ "field_tp": 0.0,
972
+ "field_fp": 0.0,
973
+ "field_fn": 0.0,
974
+ "slice_sst_pass": 0.0,
975
+ "kpi_text_presence_pass": 0.0,
976
+ "adversarial_status_pass": 0.0
977
+ },
978
+ "resume": {
979
+ "num_examples": 4,
980
+ "parse_json": 0.25,
981
+ "gold_parse_json": 1.0,
982
+ "exact_match": 0.0,
983
+ "field_precision": 0.0,
984
+ "field_recall": 0.0,
985
+ "field_f1": 0.0,
986
+ "field_tp": 0.0,
987
+ "field_fp": 1.5,
988
+ "field_fn": 1.5,
989
+ "slice_sst_pass": 0.0,
990
+ "kpi_text_presence_pass": 0.0,
991
+ "adversarial_status_pass": 0.25
992
+ },
993
+ "suspend": {
994
+ "num_examples": 1,
995
+ "parse_json": 0.0,
996
+ "gold_parse_json": 1.0,
997
+ "exact_match": 0.0,
998
+ "field_precision": 0.0,
999
+ "field_recall": 0.0,
1000
+ "field_f1": 0.0,
1001
+ "field_tp": 0.0,
1002
+ "field_fp": 0.0,
1003
+ "field_fn": 0.0,
1004
+ "slice_sst_pass": 0.0,
1005
+ "kpi_text_presence_pass": 0.0,
1006
+ "adversarial_status_pass": 0.0
1007
+ }
1008
+ }
1009
+ },
1010
+ "test_sector_ood": {
1011
+ "num_examples": 200,
1012
+ "parse_json": 0.345,
1013
+ "gold_parse_json": 1.0,
1014
+ "exact_match": 0.0,
1015
+ "field_precision": 0.0012675865800865801,
1016
+ "field_recall": 0.0008765822784810127,
1017
+ "field_f1": 0.001016076991544281,
1018
+ "field_tp": 0.04,
1019
+ "field_fp": 7.095,
1020
+ "field_fn": 12.085,
1021
+ "slice_sst_pass": 0.28,
1022
+ "kpi_text_presence_pass": 0.315,
1023
+ "adversarial_status_pass": 0.345,
1024
+ "by_target_layer": {
1025
+ "a1_policy": {
1026
+ "num_examples": 29,
1027
+ "parse_json": 0.13793103448275862,
1028
+ "gold_parse_json": 1.0,
1029
+ "exact_match": 0.0,
1030
+ "field_precision": 0.0,
1031
+ "field_recall": 0.0,
1032
+ "field_f1": 0.0,
1033
+ "field_tp": 0.0,
1034
+ "field_fp": 1.6206896551724137,
1035
+ "field_fn": 2.6206896551724137,
1036
+ "slice_sst_pass": 0.10344827586206896,
1037
+ "kpi_text_presence_pass": 0.13793103448275862,
1038
+ "adversarial_status_pass": 0.13793103448275862
1039
+ },
1040
+ "camara": {
1041
+ "num_examples": 43,
1042
+ "parse_json": 0.6046511627906976,
1043
+ "gold_parse_json": 1.0,
1044
+ "exact_match": 0.0,
1045
+ "field_precision": 0.0,
1046
+ "field_recall": 0.0,
1047
+ "field_f1": 0.0,
1048
+ "field_tp": 0.0,
1049
+ "field_fp": 4.837209302325581,
1050
+ "field_fn": 9.744186046511627,
1051
+ "slice_sst_pass": 0.5116279069767442,
1052
+ "kpi_text_presence_pass": 0.6046511627906976,
1053
+ "adversarial_status_pass": 0.6046511627906976
1054
+ },
1055
+ "etsi_zsm": {
1056
+ "num_examples": 17,
1057
+ "parse_json": 0.5882352941176471,
1058
+ "gold_parse_json": 1.0,
1059
+ "exact_match": 0.0,
1060
+ "field_precision": 0.0,
1061
+ "field_recall": 0.0,
1062
+ "field_f1": 0.0,
1063
+ "field_tp": 0.0,
1064
+ "field_fp": 20.941176470588236,
1065
+ "field_fn": 33.529411764705884,
1066
+ "slice_sst_pass": 0.5294117647058824,
1067
+ "kpi_text_presence_pass": 0.5882352941176471,
1068
+ "adversarial_status_pass": 0.5882352941176471
1069
+ },
1070
+ "intent_3gpp": {
1071
+ "num_examples": 34,
1072
+ "parse_json": 0.4411764705882353,
1073
+ "gold_parse_json": 1.0,
1074
+ "exact_match": 0.0,
1075
+ "field_precision": 0.0055147058823529415,
1076
+ "field_recall": 0.004411764705882353,
1077
+ "field_f1": 0.0049019607843137246,
1078
+ "field_tp": 0.17647058823529413,
1079
+ "field_fp": 14.823529411764707,
1080
+ "field_fn": 17.470588235294116,
1081
+ "slice_sst_pass": 0.3235294117647059,
1082
+ "kpi_text_presence_pass": 0.4411764705882353,
1083
+ "adversarial_status_pass": 0.4411764705882353
1084
+ },
1085
+ "o1_nrm": {
1086
+ "num_examples": 21,
1087
+ "parse_json": 0.19047619047619047,
1088
+ "gold_parse_json": 1.0,
1089
+ "exact_match": 0.0,
1090
+ "field_precision": 0.0,
1091
+ "field_recall": 0.0,
1092
+ "field_f1": 0.0,
1093
+ "field_tp": 0.0,
1094
+ "field_fp": 1.8095238095238095,
1095
+ "field_fn": 4.380952380952381,
1096
+ "slice_sst_pass": 0.14285714285714285,
1097
+ "kpi_text_presence_pass": 0.0,
1098
+ "adversarial_status_pass": 0.19047619047619047
1099
+ },
1100
+ "tmf921": {
1101
+ "num_examples": 51,
1102
+ "parse_json": 0.1568627450980392,
1103
+ "gold_parse_json": 1.0,
1104
+ "exact_match": 0.0,
1105
+ "field_precision": 0.001294457176810118,
1106
+ "field_recall": 0.0004964010920824026,
1107
+ "field_f1": 0.0007166417969056782,
1108
+ "field_tp": 0.0392156862745098,
1109
+ "field_fp": 4.647058823529412,
1110
+ "field_fn": 12.352941176470589,
1111
+ "slice_sst_pass": 0.13725490196078433,
1112
+ "kpi_text_presence_pass": 0.1568627450980392,
1113
+ "adversarial_status_pass": 0.1568627450980392
1114
+ },
1115
+ "tmf921_lifecycle_monitor": {
1116
+ "num_examples": 3,
1117
+ "parse_json": 0.3333333333333333,
1118
+ "gold_parse_json": 1.0,
1119
+ "exact_match": 0.0,
1120
+ "field_precision": 0.0,
1121
+ "field_recall": 0.0,
1122
+ "field_f1": 0.0,
1123
+ "field_tp": 0.0,
1124
+ "field_fp": 8.666666666666666,
1125
+ "field_fn": 10.0,
1126
+ "slice_sst_pass": 0.3333333333333333,
1127
+ "kpi_text_presence_pass": 0.0,
1128
+ "adversarial_status_pass": 0.3333333333333333
1129
+ },
1130
+ "tmf921_lifecycle_report": {
1131
+ "num_examples": 1,
1132
+ "parse_json": 0.0,
1133
+ "gold_parse_json": 1.0,
1134
+ "exact_match": 0.0,
1135
+ "field_precision": 0.0,
1136
+ "field_recall": 0.0,
1137
+ "field_f1": 0.0,
1138
+ "field_tp": 0.0,
1139
+ "field_fp": 0.0,
1140
+ "field_fn": 0.0,
1141
+ "slice_sst_pass": 0.0,
1142
+ "kpi_text_presence_pass": 0.0,
1143
+ "adversarial_status_pass": 0.0
1144
+ },
1145
+ "tmf921_lifecycle_resume": {
1146
+ "num_examples": 1,
1147
+ "parse_json": 1.0,
1148
+ "gold_parse_json": 1.0,
1149
+ "exact_match": 0.0,
1150
+ "field_precision": 0.0,
1151
+ "field_recall": 0.0,
1152
+ "field_f1": 0.0,
1153
+ "field_tp": 0.0,
1154
+ "field_fp": 3.0,
1155
+ "field_fn": 6.0,
1156
+ "slice_sst_pass": 0.0,
1157
+ "kpi_text_presence_pass": 0.0,
1158
+ "adversarial_status_pass": 1.0
1159
+ }
1160
+ },
1161
+ "by_slice_type": {
1162
+ "HMTC": {
1163
+ "num_examples": 17,
1164
+ "parse_json": 0.4117647058823529,
1165
+ "gold_parse_json": 1.0,
1166
+ "exact_match": 0.0,
1167
+ "field_precision": 0.0,
1168
+ "field_recall": 0.0,
1169
+ "field_f1": 0.0,
1170
+ "field_tp": 0.0,
1171
+ "field_fp": 8.176470588235293,
1172
+ "field_fn": 13.352941176470589,
1173
+ "slice_sst_pass": 0.35294117647058826,
1174
+ "kpi_text_presence_pass": 0.35294117647058826,
1175
+ "adversarial_status_pass": 0.4117647058823529
1176
+ },
1177
+ "MPS": {
1178
+ "num_examples": 19,
1179
+ "parse_json": 0.2631578947368421,
1180
+ "gold_parse_json": 1.0,
1181
+ "exact_match": 0.0,
1182
+ "field_precision": 0.0,
1183
+ "field_recall": 0.0,
1184
+ "field_f1": 0.0,
1185
+ "field_tp": 0.0,
1186
+ "field_fp": 4.7368421052631575,
1187
+ "field_fn": 6.947368421052632,
1188
+ "slice_sst_pass": 0.2631578947368421,
1189
+ "kpi_text_presence_pass": 0.2631578947368421,
1190
+ "adversarial_status_pass": 0.2631578947368421
1191
+ },
1192
+ "URLLC": {
1193
+ "num_examples": 58,
1194
+ "parse_json": 0.3275862068965517,
1195
+ "gold_parse_json": 1.0,
1196
+ "exact_match": 0.0,
1197
+ "field_precision": 0.0,
1198
+ "field_recall": 0.0,
1199
+ "field_f1": 0.0,
1200
+ "field_tp": 0.0,
1201
+ "field_fp": 7.362068965517241,
1202
+ "field_fn": 12.293103448275861,
1203
+ "slice_sst_pass": 0.25862068965517243,
1204
+ "kpi_text_presence_pass": 0.29310344827586204,
1205
+ "adversarial_status_pass": 0.3275862068965517
1206
+ },
1207
+ "V2X": {
1208
+ "num_examples": 19,
1209
+ "parse_json": 0.21052631578947367,
1210
+ "gold_parse_json": 1.0,
1211
+ "exact_match": 0.0,
1212
+ "field_precision": 0.0,
1213
+ "field_recall": 0.0,
1214
+ "field_f1": 0.0,
1215
+ "field_tp": 0.0,
1216
+ "field_fp": 2.526315789473684,
1217
+ "field_fn": 4.421052631578948,
1218
+ "slice_sst_pass": 0.05263157894736842,
1219
+ "kpi_text_presence_pass": 0.10526315789473684,
1220
+ "adversarial_status_pass": 0.21052631578947367
1221
+ },
1222
+ "eMBB": {
1223
+ "num_examples": 64,
1224
+ "parse_json": 0.4375,
1225
+ "gold_parse_json": 1.0,
1226
+ "exact_match": 0.0,
1227
+ "field_precision": 0.003961208062770563,
1228
+ "field_recall": 0.0027393196202531644,
1229
+ "field_f1": 0.0031752405985758783,
1230
+ "field_tp": 0.125,
1231
+ "field_fp": 8.59375,
1232
+ "field_fn": 15.1875,
1233
+ "slice_sst_pass": 0.421875,
1234
+ "kpi_text_presence_pass": 0.421875,
1235
+ "adversarial_status_pass": 0.4375
1236
+ },
1237
+ "mMTC": {
1238
+ "num_examples": 23,
1239
+ "parse_json": 0.2608695652173913,
1240
+ "gold_parse_json": 1.0,
1241
+ "exact_match": 0.0,
1242
+ "field_precision": 0.0,
1243
+ "field_recall": 0.0,
1244
+ "field_f1": 0.0,
1245
+ "field_tp": 0.0,
1246
+ "field_fp": 7.173913043478261,
1247
+ "field_fn": 12.565217391304348,
1248
+ "slice_sst_pass": 0.08695652173913043,
1249
+ "kpi_text_presence_pass": 0.2608695652173913,
1250
+ "adversarial_status_pass": 0.2608695652173913
1251
+ }
1252
+ },
1253
+ "by_lifecycle_operation": {
1254
+ "create": {
1255
+ "num_examples": 195,
1256
+ "parse_json": 0.3435897435897436,
1257
+ "gold_parse_json": 1.0,
1258
+ "exact_match": 0.0,
1259
+ "field_precision": 0.0013000888000888001,
1260
+ "field_recall": 0.0008990587471600129,
1261
+ "field_f1": 0.001042130247737724,
1262
+ "field_tp": 0.041025641025641026,
1263
+ "field_fp": 7.128205128205129,
1264
+ "field_fn": 12.21025641025641,
1265
+ "slice_sst_pass": 0.28205128205128205,
1266
+ "kpi_text_presence_pass": 0.3230769230769231,
1267
+ "adversarial_status_pass": 0.3435897435897436
1268
+ },
1269
+ "monitor": {
1270
+ "num_examples": 3,
1271
+ "parse_json": 0.3333333333333333,
1272
+ "gold_parse_json": 1.0,
1273
+ "exact_match": 0.0,
1274
+ "field_precision": 0.0,
1275
+ "field_recall": 0.0,
1276
+ "field_f1": 0.0,
1277
+ "field_tp": 0.0,
1278
+ "field_fp": 8.666666666666666,
1279
+ "field_fn": 10.0,
1280
+ "slice_sst_pass": 0.3333333333333333,
1281
+ "kpi_text_presence_pass": 0.0,
1282
+ "adversarial_status_pass": 0.3333333333333333
1283
+ },
1284
+ "report": {
1285
+ "num_examples": 1,
1286
+ "parse_json": 0.0,
1287
+ "gold_parse_json": 1.0,
1288
+ "exact_match": 0.0,
1289
+ "field_precision": 0.0,
1290
+ "field_recall": 0.0,
1291
+ "field_f1": 0.0,
1292
+ "field_tp": 0.0,
1293
+ "field_fp": 0.0,
1294
+ "field_fn": 0.0,
1295
+ "slice_sst_pass": 0.0,
1296
+ "kpi_text_presence_pass": 0.0,
1297
+ "adversarial_status_pass": 0.0
1298
+ },
1299
+ "resume": {
1300
+ "num_examples": 1,
1301
+ "parse_json": 1.0,
1302
+ "gold_parse_json": 1.0,
1303
+ "exact_match": 0.0,
1304
+ "field_precision": 0.0,
1305
+ "field_recall": 0.0,
1306
+ "field_f1": 0.0,
1307
+ "field_tp": 0.0,
1308
+ "field_fp": 3.0,
1309
+ "field_fn": 6.0,
1310
+ "slice_sst_pass": 0.0,
1311
+ "kpi_text_presence_pass": 0.0,
1312
+ "adversarial_status_pass": 1.0
1313
+ }
1314
+ }
1315
+ },
1316
+ "test_adversarial": {
1317
+ "num_examples": 33,
1318
+ "parse_json": 0.0,
1319
+ "gold_parse_json": 1.0,
1320
+ "exact_match": 0.0,
1321
+ "field_precision": 0.0,
1322
+ "field_recall": 0.0,
1323
+ "field_f1": 0.0,
1324
+ "field_tp": 0.0,
1325
+ "field_fp": 0.0,
1326
+ "field_fn": 0.0,
1327
+ "slice_sst_pass": 0.0,
1328
+ "kpi_text_presence_pass": 0.0,
1329
+ "adversarial_status_pass": 0.0,
1330
+ "by_target_layer": {
1331
+ "adversarial_ambiguous": {
1332
+ "num_examples": 17,
1333
+ "parse_json": 0.0,
1334
+ "gold_parse_json": 1.0,
1335
+ "exact_match": 0.0,
1336
+ "field_precision": 0.0,
1337
+ "field_recall": 0.0,
1338
+ "field_f1": 0.0,
1339
+ "field_tp": 0.0,
1340
+ "field_fp": 0.0,
1341
+ "field_fn": 0.0,
1342
+ "slice_sst_pass": 0.0,
1343
+ "kpi_text_presence_pass": 0.0,
1344
+ "adversarial_status_pass": 0.0
1345
+ },
1346
+ "adversarial_contradictory": {
1347
+ "num_examples": 9,
1348
+ "parse_json": 0.0,
1349
+ "gold_parse_json": 1.0,
1350
+ "exact_match": 0.0,
1351
+ "field_precision": 0.0,
1352
+ "field_recall": 0.0,
1353
+ "field_f1": 0.0,
1354
+ "field_tp": 0.0,
1355
+ "field_fp": 0.0,
1356
+ "field_fn": 0.0,
1357
+ "slice_sst_pass": 0.0,
1358
+ "kpi_text_presence_pass": 0.0,
1359
+ "adversarial_status_pass": 0.0
1360
+ },
1361
+ "adversarial_out_of_scope": {
1362
+ "num_examples": 7,
1363
+ "parse_json": 0.0,
1364
+ "gold_parse_json": 1.0,
1365
+ "exact_match": 0.0,
1366
+ "field_precision": 0.0,
1367
+ "field_recall": 0.0,
1368
+ "field_f1": 0.0,
1369
+ "field_tp": 0.0,
1370
+ "field_fp": 0.0,
1371
+ "field_fn": 0.0,
1372
+ "slice_sst_pass": 0.0,
1373
+ "kpi_text_presence_pass": 0.0,
1374
+ "adversarial_status_pass": 0.0
1375
+ }
1376
+ },
1377
+ "by_slice_type": {
1378
+ "N/A": {
1379
+ "num_examples": 33,
1380
+ "parse_json": 0.0,
1381
+ "gold_parse_json": 1.0,
1382
+ "exact_match": 0.0,
1383
+ "field_precision": 0.0,
1384
+ "field_recall": 0.0,
1385
+ "field_f1": 0.0,
1386
+ "field_tp": 0.0,
1387
+ "field_fp": 0.0,
1388
+ "field_fn": 0.0,
1389
+ "slice_sst_pass": 0.0,
1390
+ "kpi_text_presence_pass": 0.0,
1391
+ "adversarial_status_pass": 0.0
1392
+ }
1393
+ },
1394
+ "by_lifecycle_operation": {
1395
+ "create": {
1396
+ "num_examples": 33,
1397
+ "parse_json": 0.0,
1398
+ "gold_parse_json": 1.0,
1399
+ "exact_match": 0.0,
1400
+ "field_precision": 0.0,
1401
+ "field_recall": 0.0,
1402
+ "field_f1": 0.0,
1403
+ "field_tp": 0.0,
1404
+ "field_fp": 0.0,
1405
+ "field_fn": 0.0,
1406
+ "slice_sst_pass": 0.0,
1407
+ "kpi_text_presence_pass": 0.0,
1408
+ "adversarial_status_pass": 0.0
1409
+ }
1410
+ }
1411
+ }
1412
+ }
results/baselines/zero_shot_vs_finetuned.md ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Zero-shot Qwen3-8B vs Fine-tuned Qwen3-8B QLoRA
2
+
3
+ Zero-shot baseline was evaluated on 200 examples per split. Fine-tuned results are full split metrics.
4
+
5
+ | Split | Zero-shot parse | Fine-tuned parse | Zero-shot norm field F1 | Fine-tuned norm field F1 | Zero-shot norm key F1 | Fine-tuned norm key F1 |
6
+ |---|---:|---:|---:|---:|---:|---:|
7
+ | ID | 0.335 | 1.000 | 0.0009 | 0.7956 | 0.0169 | 0.9811 |
8
+ | Template OOD | 0.340 | 1.000 | 0.0014 | 0.7865 | 0.0172 | 0.9801 |
9
+ | Use-case OOD | 0.325 | 0.9998 | 0.0012 | 0.7907 | 0.0198 | 0.9805 |
10
+ | Sector OOD | 0.345 | 1.000 | 0.0008 | 0.7697 | 0.0171 | 0.9818 |
11
+ | Adversarial | 0.000 | 1.000 | 0.0000 | 0.9697 | 0.0000 | 1.0000 |
12
+
13
+ Conclusion: domain QLoRA fine-tuning is essential for structured telecom intent-to-config generation.