File size: 75,110 Bytes
e17c9f2
 
0b6004e
e17c9f2
 
0b6004e
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0b6004e
e17c9f2
 
f48c751
0b6004e
 
 
 
 
 
e17c9f2
 
 
 
 
 
b6336ac
 
 
 
 
 
 
 
e17c9f2
 
 
 
 
 
 
 
cc402d9
e17c9f2
 
 
 
 
cc402d9
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
 
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc402d9
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
 
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc402d9
e17c9f2
 
 
 
 
 
b6336ac
e17c9f2
 
 
a6a5155
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
 
 
 
 
 
 
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
 
 
e17c9f2
 
 
 
 
 
b6336ac
 
 
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
 
 
 
 
 
 
e17c9f2
 
 
 
 
b6336ac
 
e17c9f2
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
b6336ac
 
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b6336ac
e17c9f2
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
from .api import HelperCompany
import re
import os
from .header import get_dir, Prompt, ConfigReader
import traceback
import openai

TAG_moti = "Motivations:"
TAG_contr = "Details:"


def clean_text(text):
    cleaned_text = re.sub(r"-\s*\n", "", text)
    cleaned_text = re.sub(r"\s*\n\s*", " ", cleaned_text)
    return cleaned_text.strip()


def clean_entities(input_string):
    # 取出括号中的内容
    cleaned_text = re.sub(r"\([^)]*\)", "", input_string)
    # 使用正则表达式删除非字母字符
    cleaned = re.sub(r"[^a-zA-Z\s]", "", input_string)
    # 将多个空格替换为一个空格
    cleaned = re.sub(r"\s+", " ", cleaned)
    # 删除首尾空格
    cleaned = cleaned.strip().lower()
    return cleaned


class APIHelper(object):

    def __init__(self, config) -> None:
        super(APIHelper, self).__init__()
        self.config = config
        self.__checkout_config__()
        self.generator = self.get_helper()
        self.prompt = Prompt(get_dir(config.ARTICLE.summarizing_prompt))

    def get_helper(self):
        MODEL_TYPE = os.environ["MODEL_TYPE"]
        MODEL_NAME = os.environ["MODEL_NAME"]
        MODEL_API_KEY = os.environ["MODEL_API_KEY"]
        BASE_URL = os.environ["BASE_URL"]
        return HelperCompany.get()[MODEL_TYPE](
            MODEL_API_KEY, MODEL_NAME, BASE_URL, timeout=None
        )

    def __checkout_config__(self):
        pass

    def __call__(self, title: str, abstract: str, introduction: str) -> dict:
        if os.environ["MODEL_NAME"] not in [
            "glm4",
            "glm4-air",
            "qwen-max",
            "qwen-plus",
        ]:
            raise ValueError(f"Check model name...")

        if title is None or abstract is None or introduction is None:
            return None
        try:
            message = [
                self.prompt.queries[0][0](
                    title=title, abstract=abstract, introduction=introduction
                )
            ]
            response1 = self.generator.create(
                messages=message,
            )
            summary = clean_text(response1.choices[0].message.content)
            message.append({"role": "assistant", "content": summary})
            message.append(self.prompt.queries[1][0]())
            response2 = self.generator.create(
                messages=message,
            )
            detail = response2.choices[0].message.content
            motivation = clean_text(detail.split(TAG_moti)[1].split(TAG_contr)[0])
            contribution = clean_text(detail.split(TAG_contr)[1])
            result = {
                "summary": summary,
                "motivation": motivation,
                "contribution": contribution,
            }
        except Exception:
            traceback.print_exc()
            return None
        return result

    def generate_entity_list(self, abstract: str, max_num: int = 5) -> list:
        common_examples = [
            {
                "content": "This paper presents a novel approach to automatic captioning of geo-tagged images by summarizing multiple webdocuments that contain information related to an image's location. The summarizer is biased by dependency pattern models towards sentences which contain features typically provided for different scene types such as those of churches, bridges, etc. Our results show that summaries biased by dependency pattern models lead to significantly higher ROUGE scores than both n-gram language models reported in previous work and also Wikipedia baseline summaries. Summaries generated using dependency patterns also lead to more readable summaries than those generated without dependency patterns.",
                "entities": "dependency pattern models, automatic captioning",
            },
            {
                "content": "In this paper, we describe the 2015 iteration of the SemEval shared task on Sentiment Analysis in Twitter. This was the most popular sentiment analysis shared task to date with more than 40 teams participating in each of the last three years. This year's shared task competition consisted of five sentiment prediction subtasks. Two were reruns from previous years: (A) sentiment expressed by a phrase in the context of a tweet, and (B) overall sentiment of a tweet. We further included three new subtasks asking to predict (C) the sentiment towards a topic in a single tweet, (D) the overall sentiment towards a topic in a set of tweets, and (E) the degree of prior polarity of a phrase.",
                "entities": "sentiment analysis, shared task",
            },
            {
                "content": 'This paper presents two different tools which may be used as a support of speech recognition. The tool "transc" is the first one and it generates the phonetic transcription (pronunciation) of given utterance. It is based mainly on fixed rules which can be defined for Czech pronunciation but it can work also with specified list of exceptions which is defined on lexicon basis. It allows the usage of "transc" for unknown text with high probability of correct phonetic transcription generation. The second part is devoted to lexicon management tool "lexedit" which may be useful in the phase of generation of pronunciation lexicon for collected corpora. The presented tool allows editing of pronunciation, playing examples of pronunciation, comparison with reference lexicon, updating of reference lexicon, etc.',
                "entities": "speech recognition, phonetic transcription, lexicon management",
            },
            {
                "content": "Previous research applying kernel methods to natural language parsing have focussed on proposing kernels over parse trees, which are hand-crafted based on domain knowledge and computational considerations. In this paper we propose a method for defining kernels in terms of a probabilistic model of parsing. This model is then trained, so that the parameters of the probabilistic model reflect the generalizations in the training data. The method we propose then uses these trained parameters to define a kernel for reranking parse trees. In experiments, we use a neural network based statistical parser as the probabilistic model, and use the resulting kernel with the Voted Perceptron algorithm to rerank the top 20 parses from the probabilistic model. This method achieves a significant improvement over the accuracy of the probabilistic model.",
                "entities": "parse trees, probabilistic model, natural language parsing",
            },
        ]

        few_shot_examples = []
        for example in common_examples:
            few_shot_examples.append(example)

        prompt_template_entity = """
        ### Task Description:
        You are an AI researcher tasked with extracting the key entities from a given research paper content. These entities should represent the most important keywords or phrases that summarize the main topics or concepts discussed in the content.

        ### Information Provided:
        **Content**: Focus on this content, and extract entities that serve as concrete manifestations of the main themes and topics within it.

        ### Approach:
        Your entity extraction should be systematic:
        - **Step 1**: Carefully read through the content to fully understand its main themes and topics.
        - **Step 2**: Identify and list key entities central to the content, ensuring each entity is relevant, meaningful, and accurately represents the content.

        ### Entity Guidelines:
        - Each entity should be no longer than 5 words and contain at least 2 words.
        - The entities should be nouns or noun phrases.
        - The total number of entities should be less than or equal to {max_num}.

        ### Examples:
        {examples}

        ### Specific information:
        I will provide you with specific information now, please use them according to the instructions above:
        **Content**: {content}

        ### Format for Your Response:
        Please just give me the entities and spilt them by ",":
        <entity 1>,<entity2>,...
        """

        if abstract is None:
            return None
        try:
            examples_str = "\n".join(
                f"[content]: {example['content']}\n[entity]: {example['entities']}\n###\n"
                for example in few_shot_examples
            )
            system_input = "Now you are an expert in extracting key entities from research contents. You are good at identifying the most important keywords or phrases that summarize the main topics or concepts discussed in the content."
            message = []
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_entity.format(
                examples=examples_str, content=abstract, max_num=str(max_num)
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            entities = response.choices[0].message.content
            entity_list = entities.strip().split(", ")
            clean_entity_list = []
            for entity in entity_list:
                entity = clean_entities(entity)
                if len(entity.split()) <= 20:
                    clean_entity_list.append(entity)

            if "entity" not in abstract.lower() and "entities" not in abstract.lower():
                clean_entity_list = [
                    re.sub(
                        r"\bentity\b|\bentities\b", "", e, flags=re.IGNORECASE
                    ).strip()
                    for e in clean_entity_list
                ]
                clean_entity_list = [e for e in clean_entity_list if e]
                clean_entity_list = [clean_entities(e) for e in clean_entity_list]
        except Exception:
            traceback.print_exc()
            return None
        return clean_entity_list

    def generate_brainstorm(self, background: str) -> str:
        prompt_template_brainstorming = """
        ### Task Description:
        You are an AI researcher tasked with brainstorming initial, innovative ideas to address a given research problem in AI. Focus on generating diverse and creative approaches rather than finalized methods. The ideas can be rough and in their infancy but should cover a range of possible directions that could be explored further.

        ### Information Provided:
        - **Research Background**: {background}

        ### Approach:
        Your brainstorming should be systematic:
        - **Step 1**: Thoroughly understand the research background.
        - **Step 2**: Generate a list of 4 to 6 high-level ideas or directions that could potentially solve problems in the given background. Be creative, think outside the box, and avoid merely rephrasing existing methods.

        ### Format for Your Response:
        Please present 4 to 6 ideas in the following format:
        **Idea 1**: [Brief description of the first idea]
        **Idea 2**: [Brief description of the second idea]
        ...
        """

        if background is None:
            print("Input background is empty ...")
            return None
        try:
            # Initial brainstorming to generate raw ideas
            brainstorming_prompt = prompt_template_brainstorming.format(
                background=background
            )
            message = []
            prompt_first = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at generating creative and original ideas."
            message.append({"role": "system", "content": prompt_first})
            message.append({"role": "user", "content": brainstorming_prompt})
            # Call the API to generate brainstorming ideas
            response_brainstorming = self.generator.create(
                messages=message,
            )
            brainstorming_ideas = response_brainstorming.choices[0].message.content

        except Exception:
            traceback.print_exc()
            return None

        return brainstorming_ideas

    def generate_problem(self, background: str, related_papers: list[dict]):
        prompt_template_problem = """
        ### Task Description:
        You will receive a research background along with summaries, backgrounds, and contributions (methods) of several related papers. Your task is to carefully analyze this information and propose a research problem that is original, clear, feasible, relevant, and significant to its field. Additionally, provide the rationales behind the proposed problem.

        ### Information Provided:
        1. **Research Background**: This is your primary focus. The research problem you propose should be a direct reflection of this background.
        2. **Related Papers**: These papers offer studies directly related to the primary research topic, providing additional insights and knowledge that will inform your proposed problem.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Begin by thoroughly understanding the core focus of the research background.
        - **Step 2**: Review the summaries, backgrounds, and contributions (methods) of the related papers to gain broader insights into the primary research topic.
        - **Step 3**: Based on the provided information, propose a research problem that meets the criteria of being original, clear, feasible, relevant, and significant. Support your problem statement with clear rationales.

        ### Specific information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research Background**: {background}
        2. **Related Papers**: {related_papers_information}

        ### Format for Your Response:
        **Research Problem**: [your problem]
        - **Rationales**: [the rationale behind your problem]
        """
        if background is None or related_papers is None:
            return None
        try:
            related_papers_information = ""
            for i, paper in enumerate(related_papers):
                related_papers_information += (
                    "Related paper {i}: ".format(i=i + 1) + paper["title"]
                )
                related_papers_information += "\nSummary: " + paper["summary"]
                related_papers_information += "\nBackgrounds: " + paper["motivation"]
                related_papers_information += (
                    "\nContributions: " + paper["contribution"] + "\n \n"
                )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at proposing novel and valuable questions based on research background."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_problem.format(
                background=background,
                related_papers_information=related_papers_information,
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            problem = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return problem, message_input

    def generate_problem_with_cue_words(
        self, background: str, related_papers: list[dict], cue_words: list
    ):
        prompt_template_problem = """
        ### Task Description:
        You will receive a research background and some cue words along with summaries, backgrounds, and contributions (methods) of several related papers. Your task is to carefully analyze this information and propose a research problem that is original, clear, feasible, relevant, and significant to its field. Additionally, provide the rationales behind the proposed problem.

        ### Information Provided:
        1. **Research Background**: This is your primary focus. The research problem you propose should be a direct reflection of this background.
        2. **Cue Words**: Some of these words can provide direction and ideas for you to ask questions. They should be the focus of your attention.
        3. **Related Papers**: These papers offer studies directly related to the primary research topic, providing additional insights and knowledge that will inform your proposed problem.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Begin by thoroughly understanding the core focus of the research background.
        - **Step 2**: Read the cue words and then determine the approximate direction of your problem.
        - **Step 3**: Review the summaries, backgrounds, and contributions (methods) of the related papers to gain broader insights into the primary research topic.
        - **Step 4**: Based on the provided information, propose a research problem that meets the criteria of being original, clear, feasible, relevant, and significant. Support your problem statement with clear rationales.

        ### Specific information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research Background**: {background}
        2. **Cue Words**: {cue_words}
        3. **Related Papers**: {related_papers_information}

        ### Format for Your Response:
        **Research Problem**: [your problem]
        - **Rationales**: [the rationale behind your problem]
        """
        if background is None or related_papers is None or cue_words is None:
            return None
        try:
            related_papers_information = ""
            for i, paper in enumerate(related_papers):
                related_papers_information += (
                    "Related paper {i}: ".format(i=i + 1) + paper["title"]
                )
                related_papers_information += "\nSummary: " + paper["summary"]
                related_papers_information += "\nBackgrounds: " + paper["motivation"]
                related_papers_information += (
                    "\nContributions: " + paper["contribution"] + "\n \n"
                )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at proposing novel and valuable questions based on research background."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_problem.format(
                background=background,
                related_papers_information=related_papers_information,
                cue_words=cue_words,
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            problem = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return problem, message_input

    def generate_inspiration(self, problem: str, related_paper: dict):
        prompt_inspiration = """
        ### Task Description:
        You will be provided with a research problem, as well as the summary, backgrounds and contributions (methods) of a related paper. Your task is to extract a novel, effective, and specific inspiration from the related paper that can help addressing the research problem, and provide a brief rationale for this inspiration.

        ### Information Provided:
        1. **Research problem**: The key issues or aspects of the problem that need to be addressed. These will serve as the foundation for generating your inspiration.
        2. **Related paper**: Draw insights from the abstract, background, and methods of the related paper. Delve deeply into these methods, understand the motivations behind them, and critically assess how they might contribute to solving the research problem. Avoid merely replicating the methods; instead, synthesize relevant aspects with your own insights to derive a novel inspiration.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly read the research problem to clearly understand the primary focus.
        - **Step 2**: Review the summary, background, and contributions (methods) of the related paper. Evaluate whether the methods proposed in the paper can provide solutions or insights relevant to the research problem.
        - **Step 3**: Based on the information provided in the paper and your own analysis, propose a novel, effective, and specific inspiration. Include a rationale explaining how this inspiration helps addressing the research problem.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research problem**: {problem}
        2. **Related paper**: {related_paper_information}

        ### Format for Your Response:
        Your output should be around 200 words and follow the format:
        **Inspiration**: [Your novel, effective, and specific idea derived from the related paper]
        - **Rationale**: [The brief reasoning behind how this inspiration help addressing the research problem]
        """
        if problem is None or related_paper is None:
            return None
        try:
            related_paper_information = ""
            related_paper_information += "Related paper : " + related_paper["title"]
            related_paper_information += "\nSummary: " + related_paper["summary"]
            related_paper_information += "\nBackgrounds: " + related_paper["motivation"]
            related_paper_information += (
                "\nContributions: " + related_paper["contribution"] + "\n \n"
            )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at extracting novel and valuable inspirations from papers."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_inspiration.format(
                problem=problem, related_paper_information=related_paper_information
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            inspiration = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return inspiration

    def generate_inspiration_with_cue_words(
        self, problem: str, related_paper: dict, cue_words: list
    ):
        prompt_inspiration = """
        ### Task Description:
        You will be provided with a research problem, some cue words as well as the summary, backgrounds and contributions (methods) of a related paper. Your task is to extract a novel, effective, and specific inspiration from the related paper that can help addressing the research problem, and provide a brief rationale for this inspiration.

        ### Information Provided:
        1. **Research problem**: The key issues or aspects of the problem that need to be addressed. These will serve as the foundation for generating your inspiration.
        2. **Cue Words**: Some of these words can provide direction for you to extract inspiration. They should be the focus of your attention.
        3. **Related paper**: Draw insights from the abstract, background, and methods of the related paper. Delve deeply into these methods, understand the motivations behind them, and critically assess how they might contribute to solving the research problem. Avoid merely replicating the methods; instead, synthesize relevant aspects with your own insights to derive a novel inspiration.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly read the research problem to clearly understand the primary focus.
        - **Step 2**: Review the summary, background, and contributions (methods) of the related paper. Evaluate whether the methods proposed in the paper can provide solutions or insights relevant to the research problem.
        - **Step 3**: Read the cue words and consider whether these words can be combined with information of the related paper to  help providing inspiration.
        - **Step 4**: Based on the information provided in the paper and your own analysis, propose a novel, effective, and specific inspiration. Include a rationale explaining how this inspiration helps addressing the research problem.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research problem**: {problem}
        2. **Cue Words**: {cue_words}
        3. **Related paper**: {related_paper_information}

        ### Format for Your Response:
        Your output should be around 200 words and follow the format:
        **Inspiration**: [Your novel, effective, and specific idea derived from the related paper]
        - **Rationale**: [The brief reasoning behind how this inspiration help addressing the research problem]
        """
        if problem is None or related_paper is None or cue_words is None:
            return None
        try:
            related_paper_information = ""
            related_paper_information += "Related paper : " + related_paper["title"]
            related_paper_information += "\nSummary: " + related_paper["summary"]
            related_paper_information += "\nBackgrounds: " + related_paper["motivation"]
            related_paper_information += (
                "\nContributions: " + related_paper["contribution"] + "\n \n"
            )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at extracting novel and valuable inspirations from papers."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_inspiration.format(
                problem=problem,
                related_paper_information=related_paper_information,
                cue_words=cue_words,
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            inspiration = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return inspiration

    def generate_idea(self, problem: str, related_papers: list[dict]) -> str:
        prompt_template_idea = """
        ### Task Description:
        You will be provided with a research problem along with its rationales. Your task is to brainstorm some ideas that are clear, innovative, valid, and comprehensive to address the problem. Additionally, some cue words along with summaries, backgrounds, and contributions (methods) of related papers will be provided as sources of inspiration for generating novel ideas.

        ### Information Provided:
        1. **Research Problem & Rationales**: The key issues or aspects of the problem that need to be addressed. These will form the foundation for generating your ideas.
        2. **Related Papers**: Draw inspiration from the abstracts, backgrounds, and methods of these papers. Delve deeply into these methods, understand the motivations behind them, and think critically about how they might inform your approach. Avoid merely stacking existing methods; instead, integrate relevant aspects with your own insights to create original solutions.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly read the research problem to understand your primary focus.
        - **Step 2**: Review the summaries, backgrounds, and contributions (methods) of the related papers to gain a broader perspective and insights relevant to the problem.
        - **Step 3**: Based on the provided information, propose some ideas that are clear, innovative, valid, and comprehensive.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research Problem & Rationales**: {problem}
        2. **Related Papers**: {related_papers_information}

        ### Format for Your Response:
        Please ensure that your final ideas include about 10 entries, presented in the following format:
        **Idea 1**: [The first method idea]
        **Idea 2**: [The second method idea]
        **Idea 3**: [The third method idea]
        ...
        """
        if problem is None or related_papers is None:
            return None
        try:
            related_papers_information = ""
            for i, dict in enumerate(related_papers):
                related_papers_information += (
                    "Related paper {i}: ".format(i=i + 1) + dict["title"]
                )
                related_papers_information += "\nSummary: " + dict["summary"]
                related_papers_information += "\nBackgrounds: " + dict["motivation"]
                related_papers_information += (
                    "\nContributions: " + dict["contribution"] + "\n \n"
                )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at using innovative and original methods to solve cutting-edge problems in the field of AI."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_idea.format(
                problem=problem, related_papers_information=related_papers_information
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea

    def generate_idea_with_cue_words(
        self, problem: str, related_papers: list[dict], cue_words: list
    ) -> str:
        prompt_template_idea = """
        ### Task Description:
        You will be provided with a research problem along with its rationales. Your task is to brainstorm some ideas that are clear, innovative, valid, and comprehensive to address the problem. Additionally, some cue words along with summaries, backgrounds, and contributions (methods) of related papers will be provided as sources of inspiration for generating novel ideas.

        ### Information Provided:
        1. **Research Problem & Rationales**: The key issues or aspects of the problem that need to be addressed. These will form the foundation for generating your ideas.
        2. **Cue Words**: Some of these words can inspire or provide direction for you to generate ideas. You can try to think deeply in these directions and perspectives.
        3. **Related Papers**: Draw inspiration from the abstracts, backgrounds, and methods of these papers. Delve deeply into these methods, understand the motivations behind them, and think critically about how they might inform your approach. Avoid merely stacking existing methods; instead, integrate relevant aspects with your own insights to create original solutions.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly read the research problem to understand your primary focus.
        - **Step 2**: Read the cue words and think about whether these words can inspire or provide direction for you to come up with ideas.
        - **Step 3**: Review the summaries, backgrounds, and contributions (methods) of the related papers to gain a broader perspective and insights relevant to the problem.
        - **Step 4**: Based on the provided information, propose some ideas that are clear, innovative, valid, and comprehensive.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research Problem & Rationales**: {problem}
        2. **Cue Words**: {cue_words}
        3. **Related Papers**: {related_papers_information}

        ### Format for Your Response:
        Please ensure that your final ideas include about 10 entries, presented in the following format:
        **Idea 1**: [The first method idea]
        **Idea 2**: [The second method idea]
        **Idea 3**: [The third method idea]
        ...
        """
        if problem is None or related_papers is None or cue_words is None:
            return None
        try:
            related_papers_information = ""
            for i, dict in enumerate(related_papers):
                related_papers_information += (
                    "Related paper {i}: ".format(i=i + 1) + dict["title"]
                )
                related_papers_information += "\nSummary: " + dict["summary"]
                related_papers_information += "\nBackgrounds: " + dict["motivation"]
                related_papers_information += (
                    "\nContributions: " + dict["contribution"] + "\n \n"
                )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at using innovative and original methods to solve cutting-edge problems in the field of AI."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_idea.format(
                problem=problem,
                related_papers_information=related_papers_information,
                cue_words=cue_words,
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea

    def generate_idea_by_inspiration(self, problem: str, inspirations: list[str]):
        prompt_template_idea = """
        ### Task Description:
        You will be provided with a research problem and its rationales, along with inspirations and their rationales extracted from related papers. Your task is to brainstorm some ideas that are clear, innovative, valid, and comprehensive to address the problem.

        ### Information Provided:
        1. **Research problem & Rationales**: The key issues or aspects of the problem that need to be addressed. These will form the foundation for generating your ideas.
        2. **Inspirations**: Insights and ideas extracted from related papers that may provide valuable perspectives or techniques applicable to the research problem.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly read and understand the research problem to identify your primary focus.
        - **Step 2**: Review the inspirations extracted from the related papers to gain a broader perspective and insights relevant to the research topic.
        - **Step 3**: Based on the provided information, propose some ideas that are clear, innovative, valid, and comprehensive.
        
        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research problem & Rationales**: {problem}
        2. **Inspirations**: {inspirations_text}

        ### Format for Your Response:
        Please ensure that your final ideas include about 10 entries, presented in the following format:
        **Idea 1**: [The first method idea]
        **Idea 2**: [The second method idea]
        **Idea 3**: [The third method idea]
        ...
        """
        if problem is None or inspirations is None:
            return None
        try:
            inspirations_text = ""
            for i, inspiration in enumerate(inspirations):
                inspirations_text += (
                    "Inspiration {i}: ".format(i=i + 1) + "\n" + inspiration + "\n \n"
                )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at using innovative and original methods to solve cutting-edge problems in the field of AI."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_idea.format(
                problem=problem, inspirations_text=inspirations_text
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea

    def generate_idea_by_inspiration_with_cue_words(
        self, problem: str, inspirations: list[str], cue_words: list
    ):
        prompt_template_idea = """
        ### Task Description:
        You will be provided with a research problem, its rationales and some cue words, along with inspirations and their rationales extracted from related papers. Your task is to brainstorm some ideas that are clear, innovative, valid, and comprehensive to address the problem.

        ### Information Provided:
        1. **Research problem & Rationales**: The key issues or aspects of the problem that need to be addressed. These will form the foundation for generating your ideas.
        2. **Cue Words**: Some of these words can inspire or provide direction for you to generate ideas. You can try to think deeply in these directions and perspectives.
        3. **Inspirations**: Insights and ideas extracted from related papers that may provide valuable perspectives or techniques applicable to the research problem.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly read and understand the research problem to identify your primary focus.
        - **Step 2**: Read the cue words and think about whether these words can inspire or provide direction for you to come up with ideas.
        - **Step 3**: Review the inspirations extracted from the related papers to gain a broader perspective and insights relevant to the research topic.
        - **Step 4**: Based on the provided information, propose some ideas that are clear, innovative, valid, and comprehensive.
        
        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research problem & Rationales**: {problem}
        2. **Cue Words**: {cue_words}
        3. **Inspirations**: {inspirations_text}

        ### Format for Your Response:
        Please ensure that your final ideas include about 10 entries, presented in the following format:
        **Idea 1**: [The first method idea]
        **Idea 2**: [The second method idea]
        **Idea 3**: [The third method idea]
        ...
        """
        if problem is None or inspirations is None or cue_words is None:
            return None
        try:
            inspirations_text = ""
            for i, inspiration in enumerate(inspirations):
                inspirations_text += (
                    "Inspiration {i}: ".format(i=i + 1) + "\n" + inspiration + "\n \n"
                )
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at using innovative and original methods to solve cutting-edge problems in the field of AI."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_idea.format(
                problem=problem,
                inspirations_text=inspirations_text,
                cue_words=cue_words,
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea

    def integrate_idea(self, background: str, brainstorm: str, idea: str) -> str:
        prompt_template_idea = """
        Task Description: 
        You will be provided with research background information along with a set of ideas you generated previously from with the related paper information, and a set of brainstorming ideas concerning the same research topic. Your task is to combine these ideas and generate new ones, the new ideas you generate should base on the ideas you generated previously, and integrate creative parts of the brainstorming ideas. Consider the background thoroughly, taking into account the novelty and practicability of each idea. If you think an idea you generate is reasonable and valuable, feel free to retain it. 

        ### Information Provided:
        1. **Research Background**: The starting point for idea generation based on the research context.
        2. **Brainstorming Ideas**: These ideas were generated purely from the research background, focusing on innovation and may not be directly related to the problem.
        3. **Generated Ideas**: These are the ideas you previously generated by considering both the research background and related papers.

        ### Approach:
        - **Step 1**: Review the research background and original ideas to understand the foundation of the problem.
        - **Step 2**: Consider the brainstorming ideas and original ideas together. Combine, improve, or expand upon them, integrating insights from the related papers.
        - **Step 3**: Propose new ideas that are innovative and practical, ensuring they align with the research background.

        ### Specific Information:
        1. **Research Background**: {background}
        2. **Brainstorming Ideas**: {brainstorm}
        3. **Generated Ideas**: {idea}

        ### Format for Your Response:
        Please ensure that your final ideas include 5-6 entries and present the integrated ideas in the following format:
        **Idea 1**: [The first method idea]
        **Idea 2**: [The second method idea]
        **Idea 3**: [The third method idea]
        ...
        """
        if background is None or brainstorm is None or idea is None:
            return None
        try:
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at generating innovative and original ideas to solve cutting-edge problems in the field of AI."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_idea.format(
                background=background, brainstorm=brainstorm, idea=idea
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea

    def filter_idea(self, idea: str, background: str) -> str:
        prompt_template_filter = """
        ### Task Description: 
        You will be provided with some ideas you previously generated, and a research background. Your task is to select 5-6 ideas that best address the problems described in the research background (priority) and ideas that are relatively novel and feasible (secondary), and then record the ideas and their content in given format. Remember that the content of idea includes everything about the idea.

        ### Information Provided:
        1. **Ideas**: These are the ideas you previously generated based on the research background and several related papers.
        2. **Research Background**: This document describes specific problems and challenges that need to be addressed.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Analyze the research background to understand the specific problems that need solutions.
        - **Step 2**: Critically review the ideas, selecting 5-6 ideas that are most effective in solving the problems in the research background (priority) and that are also relatively novel and feasible (secondary).

        ### Specific Information:
        I will provide you with specific information now; please use them according to the instructions above:
        1. **Ideas**: {idea}
        2. **Research Background**: {background}

        ### Format for Your Response:
        Please ensure that your final ideas include 5-6 entries, whose content has not been modified. Don't generate any explanation and just present the filtered ideas as well as their content in the following format:
        **Idea 1**: [The first method idea]  
        **Idea 2**: [The second method idea]  
        **Idea 3**: [The third method idea]  
        ...
        """
        if background is None or idea is None:
            return None
        try:
            message = []
            system_input = "Now you are a researcher in the field of AI. You are good at selecting the ideas that meet the requirements."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_filter.format(
                idea=idea, background=background
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea_filtered = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea_filtered

    def modify_idea(self, background: str, idea: str) -> str:
        prompt_template_modify = """
        ### Task Description:
        You will be provided with the research background and the original ideas you previously generated. Your task is to refine these original ideas by filtering out those with low feasibility and insufficient novelty while enhancing the most critical and relevant ideas to make them more novel, feasible, targeted, and specific. If applicable, you may include formulas or algorithms to support the ideas. Additionally, please adhere to the following requirements:
        1. Do not generate ideas that are repetitive or contradictory.
        2. Ensure that the generated ideas are coherent and form a cohesive whole.

        ### Information Provided:
        1. **Research background**: This is the starting point of the original idea and the basis for analyzing whether the idea should be filtered.
        2. **Original ideas**: These are the ideas you previously generated based on research background and several related papers.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Thoroughly review the research background to understand the context and objectives.
        - **Step 2**: Analyze the original ideas critically, identifying aspects with low feasibility or insufficient novelty, and then filter out them.
        - **Step 3**: Enhance the most critical and relevant ideas by making them more novel, feasible, targeted, and specific. Incorporate formulas or algorithms if they strengthen the ideas.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Research background**: {background}
        2. **Original idea**: {idea}

        ### Format for Your Response:
        Please ensure that your response only includes the final ideas, which include 2 to 4 entries, presented in the following format:
        **Idea 1**: [The first method idea]
        - **Details**: [Details of the first idea]
        **Idea 2**: [The second method idea]
        - **Details**: [Details of the second idea]
        ...
        """
        if background is None or idea is None:
            return None
        try:
            message = []
            system_input = "Now you are a researcher in the field of AI with innovative and pioneering abilities. You are good at using innovative and original methods to solve cutting-edge problems in the field of AI."
            message.append({"role": "system", "content": system_input})
            message_input = prompt_template_modify.format(
                background=background, idea=idea
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea_modified = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea_modified

    def generate_ground_truth(self, abstract: str, contribution: str, text: str) -> str:
        prompt_method = """
        ### Task Description:
        You will be provided with the abstract and a text extracted from a paper and three contributions of the paper. Your task is to filter, refine, and revise the content of the contributions through the text provided to you.

        ### Information Provided:
        1. **Abstract**: It's the abstract directly extracted from the paper.
        2. **Contributions**: These are the contributions (methods) we have summarized based on the abstract and introduction of the paper.
        3. **Text**: It's the text directly extracted from the paper, containing the methodology of the paper.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Start by reading the abstract and contributions, to understand the main work of this paper.
        - **Step 2**: Then, read the text, to find information related to the contributions and ignore other information. If you think there is missing content in the contributions section, you can add one. On the contrary, if you think there is content duplication, merge or delete one. Please ensure that the final contributions have 2 to 4 entries.
        - **Step 3**: Finally, provide specific details for each contribution as detailed and comprehensive as possible based on the content in the text. If applicable, you may include formulas or algorithms to support the ideas.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Abstract**: {abstract}
        2. **Contribution**: {contribution}
        3. **Text**: {text}

        ### Format for Your Response:
        Your output should follow the format, and please note that your subject should not be 'the paper' but 'this method' or the specific method name:
        **Idea 1**: [The first method idea]
        - **Details**: [Details of the first idea]
        **Idea 2**: [The second method idea]
        - **Details**: [Details of the second idea]
        ...
        """
        ground_truth = None
        if abstract is None or contribution is None or text is None:
            return None
        try:
            message = []
            prompt = prompt_method.format(
                abstract=abstract, contribution=contribution, text=text
            )
            message.append({"role": "user", "content": prompt})
            response = self.generator.create(
                messages=message,
            )
            ground_truth = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
        return ground_truth

    def transfer_form(self, idea: str):
        prompt_template_transfer = """
        ### Task Description:
        I will give you some ideas, please standardize the output format of the ideas without changing any characters in their content. Note that the content of each idea includes everything about the idea。

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        **Ideas**:
        '''
        {idea}
        '''
        ### Format for Your Response:
        Please ensure that your final answer is strictly presented in the following format:
        **1.**<The content of the first idea>. 
        **2.**<The content of the second idea>. 
        ...
        """
        if idea is None:
            return None
        try:
            message = []
            message_input = prompt_template_transfer.format(idea=idea)
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            idea_norm = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return idea_norm

    def select_contribution(self, idea: str, contribution: list[str]) -> str:
        prompt_template_select = """
        ### Task Description:
        You will be provided with an idea you previously generated, and some reference ideas. Your task is to select the idea that is most similar to the one you generated from the reference ideas.

        ### Information Provided:
        1. **Generated Idea**: This is the idea you previously generated based on research background and several related papers.
        2. **Reference Ideas**: These are the ideas that you should select from.

        ### Approach:
        Your approach should be systematic:
        - **Step 1**: Analyze the generated idea to understand the methods it describes.
        - **Step 2**: Critically review the reference ideas, selecting the idea that is most similar to the methods in the generated idea.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Idea**: {idea}
        2. **Reference Ideas**: {reference_ideas}

        ### Format for Your Response:
        Your answer can only have one number (strating from 1), indicating the number of the most similar idea, and cannot contain any other content.
        """
        if idea is None or contribution is None:
            return None
        try:
            message = []
            reference_ideas = ""
            for i, idea in enumerate(contribution):
                reference_ideas += "Idea {i}: ".format(i=i + 1) + "\n" + idea + "\n \n"
            message_input = prompt_template_select.format(
                idea=idea, reference_ideas=reference_ideas
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
                max_tokens=10,
            )
            index = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return index

    def get_similarity_score(self, idea: str, contribution: str) -> str:
        prompt_template_select = """
        ### Task Description:
        You will be provided with an idea you previously generated, and a reference idea. Your task is to determine the similarity between the generated idea and the reference idea and give a score from 0 to 5.

        ### Information Provided:
        1. **Generated Idea**: This is the idea you previously generated based on research background and several related papers.
        2. **Reference Idea**: This is the idea we provide you with that you need to compare with the generated idea.

        ### Approach:
        You should follow the following scoring criteria:
        - **0**: The generated idea and reference idea are completely unrelated with no discernible similarities.
        - **1**: The generated idea and reference idea have a vague connection, but differ significantly in their main concepts or approach.
        - **2**: The generated idea and reference idea share a general concept but differ in most key aspects such as methodology or application.
        - **3**: The generated idea and reference idea are similar in several areas, including general concept and some aspects of methodology, but differ in details or specific approaches.
        - **4**: The generated idea and reference idea are largely similar in concept, methodology, and approach, with only minor differences in specifics.
        - **5**: The generated idea and reference idea are nearly identical in all key aspects, including concept, methodology, and approach.

        ### Specific Information:
        I will provide you with specific information now, please use them according to the instructions above:
        1. **Generated Idea**: {idea}
        2. **Reference Idea**: {reference_idea}

        ### Format for Your Response:
        Your answer can only have one number (from 0 to 5), indicating the similarity score, and cannot contain any other content.
        """
        if idea is None or contribution is None:
            return None
        try:
            message = []
            reference_ideas = ""
            message_input = prompt_template_select.format(
                idea=idea, reference_idea=contribution
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
                max_tokens=10,
            )
            score = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return score

    def novelty_eval(
        self,
        current_round: int,
        num_rounds: int,
        max_num_iterations: int,
        idea: str,
        last_query_results: str,
        msg_history: list,
    ):
        novelty_system_msg = """You are an ambitious AI PhD student who is looking to publish a paper that will contribute significantly to the field.
        You have an idea and you want to check if it is novel or not. I.e., not overlapping significantly with existing literature or already well explored.
        Be a harsh critic for novelty, ensure there is a sufficient contribution in the idea for a new conference or workshop paper.
        You will be given access to the Semantic Scholar API, which you may use to survey the literature and find relevant papers to help you make your decision.
        The top 10 results for any search query will be presented to you with the abstracts.

        You will be given {num_rounds} rounds to decide on the paper.
        At any round, compare the provided idea with the information found in the article and provide a novelty score from 0 to 10.
        In each search round, you should give a query and a novelty score based on the information in the relevant papers. 
        If there are no relevant papers, give a novelty score based on your own feelings.
        """

        novelty_prompt = '''Round {current_round}/{num_rounds}.
        You have this idea:

        """
        {idea}
        """

        The results of the last query are (empty on first round):
        """
        {last_query_results}
        """

        Respond in the following format:

        THOUGHT:
        <THOUGHT>

        RESPONSE:
        ```json
        <JSON>
        ```

        In <THOUGHT>, first briefly reason over the idea and identify any query that could help you suggest a score based on its novelty. Then give your perceived novelty score.

        In <JSON>, respond in JSON format with ONLY the following field:
        - "Query": An optional search query to search the literature (e.g. attention is all you need). You must make a query if you have not decided this round.
        - "Novelty Score": A novelty score from 0 to 10.

        A query will work best if you are able to recall the exact name of the paper you are looking for, or the authors.
        This JSON will be automatically parsed, so ensure the format is precise. (the JSON MUST contain the "Query" and the "Novelty Score")
        In the last round, you should assign a "" value to the "Query" even if you don't need to generate it.'''
        msg = novelty_prompt.format(
            current_round=current_round,
            num_rounds=max_num_iterations,
            idea=idea,
            last_query_results=last_query_results,
        )
        system_message = novelty_system_msg.format(
            num_rounds=max_num_iterations,
        )
        if msg_history is None:
            msg_history = []
        try:
            new_msg_history = msg_history + [{"role": "user", "content": msg}]
            response = self.generator.create(
                messages=[
                    {"role": "system", "content": system_message},
                    *new_msg_history,
                ],
                temperature=0.75,
                max_tokens=3000,
                n=1,
                stop=None,
                seed=0,
            )
            content = response.choices[0].message.content
            new_msg_history = new_msg_history + [
                {"role": "assistant", "content": content}
            ]

        except Exception:
            traceback.print_exc()
            return None
        return content, new_msg_history

    def compare_same(
        self, idea1: str, idea2: str, idea3: str, idea4: str, idea5: str
    ) -> str:
        system_input = """
        You are an artificial intelligence researcher with extensive knowledge in this field, and now you need to make a comprehensive comparison among five ideas.
        You will obtain a comparison standard, compare every point on the standard.
        """
        input_message = '''
        ### Comparison Standard:
        """
        **Clarity**: It evaluates whether the method is articulated in a straightforward and coherent manner, facilitating a comprehensive understanding for both practitioners and researchers, thus enabling effective application and potential adaptation in similar studies.
        **Novelty**: It assesses the degree to which the method presents novel ideas or transformative strategies that challenge conventional practices, fostering advancements in the field and inspiring future research directions.
        **Feasibility**: It examines the practicality and implementability of the method, ensuring that the required resources, time, and expertise are realistically available for its execution within the constraints of the study environment.
        **Generalizability**: It determines how broadly the method can be extended or adapted to various contexts, populations, or situations, evaluating its applicability beyond the specific conditions of the study while maintaining relevance and effectiveness.
        """

        ### You should compare these five ideas:
        """IDEA1
        {idea1}
        """
        """IDEA2
        {idea2}
        """
        """IDEA3
        {idea3}
        """
        """IDEA4
        {idea4}
        """
        """IDEA5
        {idea5}
        """

        ### Respond in the following format:

        THOUGHT:
        ```thought
        <THOUGHT>
        ```

        RESPONSE:
        ```json
        <JSON>
        ```

        In <THOUGHT>, You can record your reasoning process to make your comparison more organized..

        In <JSON>, respond in JSON format with ONLY the following field:
        - "Clarity": Provide an array consisting of 1-5, representing each idea separately, with the better idea placed at the beginning (e.g. [4, 5, 3, 2, 1])
        - "Novelty": Same as above.
        - "Feasibility": Same as above.
        - "Generalizability": Same as above.
        - "Overall Ranking": Same as above.

        This JSON will be automatically parsed, so ensure the format is precise.
        '''
        if (
            idea1 is None
            or idea2 is None
            or idea3 is None
            or idea4 is None
            or idea5 is None
        ):
            return None
        try:
            message = []
            message.append({"role": "system", "content": system_input})
            message_input = input_message.format(
                idea1=idea1, idea2=idea2, idea3=idea3, idea4=idea4, idea5=idea5
            )
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            result = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return result

    def compare_all(self, idea1: str, idea2: str) -> str:
        system_input = """
        You are an artificial intelligence researcher with extensive knowledge in this field, and now you need to make a comprehensive comparison among five ideas.
        You will obtain a comparison standard, compare every point on the standard, and make a overall ranking at the end.
        """
        input_message = '''
        ### Comparison Standard:
        """
        **Novelty**: It assesses the degree to which the method presents novel ideas or transformative strategies that challenge conventional practices, fostering advancements in the field and inspiring future research directions.
        **Feasibility**: It examines the practicality and implementability of the method, ensuring that the required resources, time, and expertise are realistically available for its execution within the constraints of the study environment.
        **Clarity**: It evaluates whether the method is articulated in a straightforward and coherent manner, facilitating a comprehensive understanding for both practitioners and researchers, thus enabling effective application and potential adaptation in similar studies.
        **Generalizability**: It determines how broadly the method can be extended or adapted to various contexts, populations, or situations, evaluating its applicability beyond the specific conditions of the study while maintaining relevance and effectiveness.
        """

        ### You should compare these five ideas:
        """IDEA1
        {idea1}
        """
        """IDEA2
        {idea2}
        """

        ### Respond in the following format:

        THOUGHT:
        ```thought
        <THOUGHT>
        ```

        RESPONSE:
        ```json
        <JSON>
        ```

        In <THOUGHT>, You can record your reasoning process and explain why you think the idea is better in each aspect in detail to make your comparison more organized.

        In <JSON>, respond in JSON format with ONLY the following field:
        - "Novelty": Provide an array consisting of 1 and 2, representing each idea separately, with the better idea placed at the beginning (e.g. [1, 2]).
        - "Feasibility": Same as above.
        - "clarity": Same as above.
        - "Generalizability": Same as above.
        - "Overall Ranking": Same as above.


        This THOUGHT and JSON will be automatically parsed, so ensure the format is precise.
        '''
        if idea1 is None or idea2 is None:
            return None
        try:
            message = []
            message.append({"role": "system", "content": system_input})
            message_input = input_message.format(idea1=idea1, idea2=idea2)
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            result = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return result

    def compare_novelty_and_feasibility(self, idea1: str, idea2: str) -> str:
        system_input = """
        You are an artificial intelligence researcher with extensive knowledge in this field, and now you need to make a comprehensive comparison between two ideas.
        You will obtain a comparison standard, compare every point on the standard, and make a ranking at the end.
        """
        input_message = '''
        ### Comparison Standard:
        """
        **Novelty**: It assesses the degree to which the method presents novel ideas or transformative strategies that challenge conventional practices, fostering advancements in the field and inspiring future research directions.
        **Feasibility**: It examines the practicality and implementability of the method, ensuring that the required resources, time, and expertise are realistically available for its execution within the constraints of the study environment.
        """

        ### You should compare these five ideas:
        """IDEA1
        {idea1}
        """
        """IDEA2
        {idea2}
        """

        ### Respond in the following format:

        THOUGHT:
        ```thought
        <THOUGHT>
        ```

        RESPONSE:
        ```json
        <JSON>
        ```

        In <THOUGHT>, You can record your reasoning process and explain why you think the idea is better in each aspect in detail to make your comparison more organized.

        In <JSON>, respond in JSON format with ONLY the following field:
        - "Novelty": Provide an array consisting of 1 and 2, representing each idea separately, with the better idea placed at the beginning (e.g. [1, 2]).
        - "Feasibility": Same as above.

        This THOUGHT and JSON will be automatically parsed, so ensure the format is precise.
        '''
        if idea1 is None or idea2 is None:
            return None
        try:
            message = []
            message.append({"role": "system", "content": system_input})
            message_input = input_message.format(idea1=idea1, idea2=idea2)
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            result = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return result

    def compare_novelty(self, idea1: str, idea2: str) -> str:
        system_input = """
        You are an artificial intelligence researcher with extensive knowledge in this field, and now you need to make a comparison between two ideas.
        You will obtain a comparison standard, compare the novelty between the ideas, and make a ranking at the end.
        """
        input_message = '''
        ### Comparison Standard:
        """
        **Novelty**: It assesses the degree to which the method presents novel ideas or transformative strategies that challenge conventional practices, fostering advancements in the field and inspiring future research directions.
        """

        ### You should compare these five ideas:
        """IDEA1
        {idea1}
        """
        """IDEA2
        {idea2}
        """

        ### Respond in the following format:

        THOUGHT:
        ```thought
        <THOUGHT>
        ```

        RESPONSE:
        ```json
        <JSON>
        ```

        In <THOUGHT>, You can record your reasoning process and explain why you think the idea is better in each aspect in detail to make your comparison more organized.

        In <JSON>, respond in JSON format with ONLY the following field:
        - "Novelty": Provide an array consisting of 1 and 2, representing each idea separately, with the better idea placed at the beginning (e.g. [1, 2]).

        This THOUGHT and JSON will be automatically parsed, so ensure the format is precise and don't forget the label "Novelty".
        '''
        if idea1 is None or idea2 is None:
            return None
        try:
            message = []
            message.append({"role": "system", "content": system_input})
            message_input = input_message.format(idea1=idea1, idea2=idea2)
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            result = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return result

    def compare_feasibility(self, idea1: str, idea2: str) -> str:
        system_input = """
        You are an artificial intelligence researcher with extensive knowledge in this field, and now you need to make a comparison between two ideas.
        You will obtain a comparison standard, compare the feasibility between the ideas, and make a ranking at the end.
        """
        input_message = '''
        ### Comparison Standard:
        """
        **Feasibility**: It examines the practicality and implementability of the method, ensuring that the required resources, time, and expertise are realistically available for its execution within the constraints of the study environment.
        """

        ### You should compare these five ideas:
        """IDEA1
        {idea1}
        """
        """IDEA2
        {idea2}
        """

        ### Respond in the following format:

        THOUGHT:
        ```thought
        <THOUGHT>
        ```

        RESPONSE:
        ```json
        <JSON>
        ```

        In <THOUGHT>, You can record your reasoning process and explain why you think the idea is better in each aspect in detail to make your comparison more organized.

        In <JSON>, respond in JSON format with ONLY the following field:
        - "Feasibility": Provide an array consisting of 1 and 2, representing each idea separately, with the better idea placed at the beginning (e.g. [1, 2]).

        This THOUGHT and JSON will be automatically parsed, so ensure the format is precise and don't forget the label "Feasibility".
        '''
        if idea1 is None or idea2 is None:
            return None
        try:
            message = []
            message.append({"role": "system", "content": system_input})
            message_input = input_message.format(idea1=idea1, idea2=idea2)
            message.append({"role": "user", "content": message_input})
            response = self.generator.create(
                messages=message,
            )
            result = response.choices[0].message.content
        except Exception:
            traceback.print_exc()
            return None
        return result


if __name__ == "__main__":
    config = ConfigReader.load("/mnt/llms/data/scimon-plus-data/configs/datasets.yaml")
    api_helper = APIHelper(config=config)