File size: 124,683 Bytes
6fa4bc9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
{
    "paper_id": "D07-1007",
    "header": {
        "generated_with": "S2ORC 1.0.0",
        "date_generated": "2023-01-19T16:19:44.233879Z"
    },
    "title": "Improving Statistical Machine Translation using Word Sense Disambiguation",
    "authors": [
        {
            "first": "Marine",
            "middle": [],
            "last": "Carpuat",
            "suffix": "",
            "affiliation": {
                "laboratory": "",
                "institution": "University of Science and Technology",
                "location": {
                    "addrLine": "Clear Water Bay",
                    "settlement": "Hong Kong"
                }
            },
            "email": "marine@cs.ust.hk"
        },
        {
            "first": "Dekai",
            "middle": [],
            "last": "Wu",
            "suffix": "",
            "affiliation": {
                "laboratory": "",
                "institution": "University of Science and Technology",
                "location": {
                    "addrLine": "Clear Water Bay",
                    "settlement": "Hong Kong"
                }
            },
            "email": ""
        }
    ],
    "year": "",
    "venue": null,
    "identifiers": {},
    "abstract": "We show for the first time that incorporating the predictions of a word sense disambiguation system within a typical phrase-based statistical machine translation (SMT) model consistently improves translation quality across all three different IWSLT Chinese-English test sets, as well as producing statistically significant improvements on the larger NIST Chinese-English MT taskand moreover never hurts performance on any test set, according not only to BLEU but to all eight most commonly used automatic evaluation metrics. Recent work has challenged the assumption that word sense disambiguation (WSD) systems are useful for SMT. Yet SMT translation quality still obviously suffers from inaccurate lexical choice. In this paper, we address this problem by investigating a new strategy for integrating WSD into an SMT system, that performs fully phrasal multi-word disambiguation. Instead of directly incorporating a Senseval-style WSD system, we redefine the WSD task to match the exact same phrasal translation disambiguation task faced by phrase-based SMT systems. Our results provide the first known empirical evidence that lexical semantics are indeed useful for SMT, despite claims to the contrary.",
    "pdf_parse": {
        "paper_id": "D07-1007",
        "_pdf_hash": "",
        "abstract": [
            {
                "text": "We show for the first time that incorporating the predictions of a word sense disambiguation system within a typical phrase-based statistical machine translation (SMT) model consistently improves translation quality across all three different IWSLT Chinese-English test sets, as well as producing statistically significant improvements on the larger NIST Chinese-English MT taskand moreover never hurts performance on any test set, according not only to BLEU but to all eight most commonly used automatic evaluation metrics. Recent work has challenged the assumption that word sense disambiguation (WSD) systems are useful for SMT. Yet SMT translation quality still obviously suffers from inaccurate lexical choice. In this paper, we address this problem by investigating a new strategy for integrating WSD into an SMT system, that performs fully phrasal multi-word disambiguation. Instead of directly incorporating a Senseval-style WSD system, we redefine the WSD task to match the exact same phrasal translation disambiguation task faced by phrase-based SMT systems. Our results provide the first known empirical evidence that lexical semantics are indeed useful for SMT, despite claims to the contrary.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Abstract",
                "sec_num": null
            }
        ],
        "body_text": [
            {
                "text": "Common assumptions about the role and usefulness of word sense disambiguation (WSD) models in full-scale statistical machine translation (SMT) systems have recently been challenged.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "On the one hand, in previous work (Carpuat and Wu, 2005b) we obtained disappointing results when using the predictions of a Senseval WSD system in conjunction with a standard word-based SMT system: we reported slightly lower BLEU scores despite trying to incorporate WSD using a number of apparently sensible methods. These results cast doubt on the assumption that sophisticated dedicated WSD systems that were developed independently from any particular NLP application can easily be integrated into a SMT system so as to improve translation quality through stronger models of context and rich linguistic information. Rather, it has been argued, SMT systems have managed to achieve significant improvements in translation quality without directly addressing translation disambiguation as a WSD task. Instead, translation disambiguation decisions are made indirectly, typically using only word surface forms and very local contextual information, forgoing the much richer linguistic information that WSD systems typically take advantage of.",
                "cite_spans": [
                    {
                        "start": 34,
                        "end": 57,
                        "text": "(Carpuat and Wu, 2005b)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "On the other hand, error analysis reveals that the performance of SMT systems still suffers from inaccurate lexical choice. In subsequent empirical studies, we have shown that SMT systems perform much worse than dedicated WSD models, both supervised and unsupervised, on a Senseval WSD task (Carpuat and Wu, 2005a) , and therefore suggest that WSD should have a role to play in state-of-the-art SMT systems. In addition to the Senseval shared tasks, which have provided standard sense inventories and data sets, WSD research has also turned increasingly to designing specific models for a particular application. For instance, Vickrey et al. (2005) and Specia (2006) proposed WSD systems designed for French to English, and Portuguese to English translation respectively, and present a more optimistic outlook for the use of WSD in MT, although these WSD systems have not yet been integrated nor evaluated in full-scale machine translation systems.",
                "cite_spans": [
                    {
                        "start": 291,
                        "end": 314,
                        "text": "(Carpuat and Wu, 2005a)",
                        "ref_id": null
                    },
                    {
                        "start": 627,
                        "end": 648,
                        "text": "Vickrey et al. (2005)",
                        "ref_id": "BIBREF38"
                    },
                    {
                        "start": 653,
                        "end": 666,
                        "text": "Specia (2006)",
                        "ref_id": "BIBREF35"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "Taken together, these seemingly contradictory results suggest that improving SMT lexical choice accuracy remains a key challenge to improve current SMT quality, and that it is still unclear what is the most appropriate integration framework for the WSD models in SMT.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "In this paper, we present first results with a new architecture that integrates a state-of-the-art WSD model into phrase-based SMT so as to perform multi-word phrasal lexical disambiguation, and show that this new WSD approach not only produces gains across all available Chinese-English IWSLT06 test sets for all eight commonly used automated MT evaluation metrics, but also produces statistically significant gains on the much larger NIST Chinese-English task. The main difference between this approach and several of our earlier approaches as described in Carpuat and Wu (2005b) and subsequently Carpuat et al. (2006) lies in the fact that we focus on repurposing the WSD system for multi-word phrase-based SMT. Rather than using a generic Senseval WSD model as we did in Carpuat and Wu (2005b) , here both the WSD training and the WSD predictions are integrated into the phrase-based SMT framework. Furthermore, rather than using a single word based WSD approach to augment a phrase-based SMT model as we did in Carpuat et al. (2006) to improve BLEU and NIST scores, here the WSD training and predictions operate on full multi-word phrasal units, resulting in significantly more reliable and consistent gains as evaluted by many other translation accuracy metrics as well. Specifically:",
                "cite_spans": [
                    {
                        "start": 559,
                        "end": 581,
                        "text": "Carpuat and Wu (2005b)",
                        "ref_id": null
                    },
                    {
                        "start": 599,
                        "end": 620,
                        "text": "Carpuat et al. (2006)",
                        "ref_id": "BIBREF7"
                    },
                    {
                        "start": 775,
                        "end": 797,
                        "text": "Carpuat and Wu (2005b)",
                        "ref_id": null
                    },
                    {
                        "start": 1016,
                        "end": 1037,
                        "text": "Carpuat et al. (2006)",
                        "ref_id": "BIBREF7"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "\u2022 Instead of using a Senseval system, we redefine the WSD task to be exactly the same as lexical choice task faced by the multi-word phrasal translation disambiguation task faced by the phrase-based SMT system.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "\u2022 Instead of using predefined senses drawn from manually constructed sense inventories such as HowNet (Dong, 1998) , our WSD for SMT system directly disambiguates between all phrasal translation candidates seen during SMT training.",
                "cite_spans": [
                    {
                        "start": 102,
                        "end": 114,
                        "text": "(Dong, 1998)",
                        "ref_id": "BIBREF13"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "\u2022 Instead of learning from manually annotated training data, our WSD system is trained on the same corpora as the SMT system.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "However, despite these adaptations to the SMT task, the core sense disambiguation task remains pure WSD:",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "\u2022 The rich context features are typical of WSD and almost never used in SMT.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "\u2022 The dynamic integration of context-sensitive translation probabilities is not typical of SMT.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "\u2022 Although it is embedded in a real SMT system, the WSD task is exactly the same as in recent and coming Senseval Multilingual Lexical Sample tasks (e.g., ), where sense inventories represent the semantic distinctions made by another language.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "We begin by presenting the WSD module and the SMT integration technique. We then show that incorporating it into a standard phrase-based SMT baseline system consistently improves translation quality across all three different test sets from the Chinese-English IWSLT text translation evaluation, as well as on the larger NIST Chinese-English translation task. Depending on the metric, the individual gains are sometimes modest, but remarkably, incorporating WSD never hurts, and helps enough to always make it a worthwile additional component in an SMT system. Finally, we analyze the reasons for the improvement.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "2 Problems in context-sensitive lexical choice for SMT",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "To the best of our knowledge, there has been no previous attempt at integrating a state-of-the-art WSD system for fully phrasal multi-word lexical choice into phrase-based SMT, with evaluation of the resulting system on a translation task. While there are many evaluations of WSD quality, in particular the Senseval series of shared tasks (Kilgarriff and Rosenzweig (1999) , Kilgarriff (2001) , ), very little work has been done to address the actual integration of WSD in realistic SMT applications.",
                "cite_spans": [
                    {
                        "start": 339,
                        "end": 372,
                        "text": "(Kilgarriff and Rosenzweig (1999)",
                        "ref_id": "BIBREF18"
                    },
                    {
                        "start": 375,
                        "end": 392,
                        "text": "Kilgarriff (2001)",
                        "ref_id": "BIBREF19"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "To fully integrate WSD into phrase-based SMT, it is necessary to perform lexical disambiguation on multi-word phrasal lexical units; in contrast, the model reported in Cabezas and Resnik (2005) can only perform lexical disambiguation on single words. Like the model proposed in this paper, Cabezas and Resnik attempted to integrate phrasebased WSD models into decoding. However, although they reported that incorporating these predictions via the Pharaoh XML markup scheme yielded a small improvement in BLEU score over a Pharaoh baseline on a single Spanish-English translation data set, we have determined empirically that applying their single-word based model to several Chinese-English datasets does not yield systematic improvements on most MT evaluation metrics (Carpuat and Wu, 2007) . The single-word model has the disadvantage of forcing the decoder to choose between the baseline phrasal translation probabilities versus the WSD model predictions for single words. In addition, the single-word model does not generalize to WSD for phrasal lexical choice, as overlapping spans cannot be specified with the XML markup scheme. Providing WSD predictions for phrases would require committing to a phrase segmentation of the input sentence before decoding, which is likely to hurt translation quality.",
                "cite_spans": [
                    {
                        "start": 168,
                        "end": 193,
                        "text": "Cabezas and Resnik (2005)",
                        "ref_id": "BIBREF2"
                    },
                    {
                        "start": 769,
                        "end": 791,
                        "text": "(Carpuat and Wu, 2007)",
                        "ref_id": "BIBREF5"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "It is also necessary to focus directly on translation accuracy rather than other measures such as alignment error rate, which may not actually lead to improved translation quality; in contrast, for example, Garcia-Varea et al. (2001) and Garcia-Varea et al. (2002) show improved alignment error rate with a maximum entropy based context-dependent lexical choice model, but not improved translation accuracy. In contrast, our evaluation in this paper is conducted on the actual decoding task, rather than intermediate tasks such as word alignment. Moreover, in the present work, all commonly available automated MT evaluation metrics are used, rather than only BLEU score, so as to maintain a more balanced perspective.",
                "cite_spans": [
                    {
                        "start": 207,
                        "end": 233,
                        "text": "Garcia-Varea et al. (2001)",
                        "ref_id": "BIBREF15"
                    },
                    {
                        "start": 238,
                        "end": 264,
                        "text": "Garcia-Varea et al. (2002)",
                        "ref_id": "BIBREF16"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "Another problem in the context-sensitive lexical choice in SMT models of Garcia Varea et al. is that their feature set is insufficiently rich to make much better predictions than the SMT model itself. In contrast, our WSD-based lexical choice models are designed to directly model the lexical choice in the actual translation direction, and take full advantage of not residing strictly within the Bayesian sourcechannel model in order to benefit from the much richer Senseval-style feature set this facilitates.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "Garcia Varea et al. found that the best results are obtained when the training of the context-dependent translation model is fully incorporated with the EM training of the SMT system. As described below, the training of our new WSD model, though not incorporated within the EM training, is also far more closely tied to the SMT model than is the case with traditional standalone WSD models.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "In contrast with Brown et al. (1991) , our approach incorporates the predictions of state-of-theart WSD models that use rich contextual features for any phrase in the input vocabulary. In Brown et al.'s early study of WSD impact on SMT performance, the authors reported improved translation quality on a French to English task, by choosing an English translation for a French word based on the single contextual feature which is reliably discriminative. However, this was a pilot study, which is limited to words with exactly two translation candidates, and it is not clear that the conclusions would generalize to more recent SMT architectures.",
                "cite_spans": [
                    {
                        "start": 17,
                        "end": 36,
                        "text": "Brown et al. (1991)",
                        "ref_id": "BIBREF1"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1"
            },
            {
                "text": "The close relationship between WSD and SMT has been emphasized since the emergence of WSD as an independent task. However, most of previous research has focused on using multilingual resources typically used in SMT systems to improve WSD accuracy, e.g., Dagan and Itai (1994) , Li and Li (2002) , Diab (2004) . In contrast, this paper focuses on the converse goal of using WSD models to improve actual translation quality.",
                "cite_spans": [
                    {
                        "start": 254,
                        "end": 275,
                        "text": "Dagan and Itai (1994)",
                        "ref_id": "BIBREF10"
                    },
                    {
                        "start": 278,
                        "end": 294,
                        "text": "Li and Li (2002)",
                        "ref_id": "BIBREF24"
                    },
                    {
                        "start": 297,
                        "end": 308,
                        "text": "Diab (2004)",
                        "ref_id": "BIBREF11"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Problems in translation-oriented WSD",
                "sec_num": "3"
            },
            {
                "text": "Recently, several researchers have focused on designing WSD systems for the specific purpose of translation. Vickrey et al. (2005) train a logistic regression WSD model on data extracted from automatically word aligned parallel corpora, but evaluate on a blank filling task, which is essentially an evaluation of WSD accuracy. Specia (2006) describes an inductive logic programming-based WSD system, which was specifically designed for the purpose of Portuguese to English translation, but this system was also only evaluated on WSD accuracy, and not integrated in a full-scale machine translation system. Ng et al. (2003) show that it is possible to use automatically word aligned parallel corpora to train accurate supervised WSD models. The purpose of the study was to lower the annotation cost for supervised WSD, as suggested earlier by Resnik and Yarowsky (1999) . However this result is also encouraging for the integration of WSD in SMT, since it suggests that accurate WSD can be achieved using training data of the kind needed for SMT.",
                "cite_spans": [
                    {
                        "start": 109,
                        "end": 130,
                        "text": "Vickrey et al. (2005)",
                        "ref_id": "BIBREF38"
                    },
                    {
                        "start": 327,
                        "end": 340,
                        "text": "Specia (2006)",
                        "ref_id": "BIBREF35"
                    },
                    {
                        "start": 606,
                        "end": 622,
                        "text": "Ng et al. (2003)",
                        "ref_id": "BIBREF26"
                    },
                    {
                        "start": 842,
                        "end": 868,
                        "text": "Resnik and Yarowsky (1999)",
                        "ref_id": "BIBREF31"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Problems in translation-oriented WSD",
                "sec_num": "3"
            },
            {
                "text": "4 Building WSD models for phrase-based SMT 4.1 WSD models for every phrase in the input vocabulary Just like for the baseline phrase translation model, WSD models are defined for every phrase in the input vocabulary. Lexical choice in SMT is naturally framed as a WSD problem, so the first step of integration consists of defining a WSD model for every phrase in the SMT input vocabulary. This differs from traditional WSD tasks, where the WSD target is a single content word. Senseval for instance has either lexical sample or all word tasks. The target words for both categories of Senseval WSD tasks are typically only content wordsprimarily nouns, verbs, and adjectives-while in the context of SMT, we need to translate entire sentences, and therefore have a WSD model not only for every word in the input sentences, regardless of their POS tag, but for every phrase, including tokens such as articles, prepositions and even punctuation. Further empirical studies have suggested that includ-ing WSD predictions for those longer phrases is a key factor to help the decoder produce better translations (Carpuat and Wu, 2007) .",
                "cite_spans": [
                    {
                        "start": 1104,
                        "end": 1126,
                        "text": "(Carpuat and Wu, 2007)",
                        "ref_id": "BIBREF5"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Problems in translation-oriented WSD",
                "sec_num": "3"
            },
            {
                "text": "Instead of using pre-defined sense inventories, the WSD models disambiguate between the SMT translation candidates. In order to closely integrate WSD predictions into the SMT system, we need to formulate WSD models so that they produce features that can directly be used in translation decisions taken by the SMT system. It is therefore necessary for the WSD and SMT systems to consider exactly the same translation candidates for a given word in the input language.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "WSD uses the same sense definitions as the SMT system",
                "sec_num": "4.2"
            },
            {
                "text": "Assuming a standard phrase-based SMT system (e.g., Koehn et al. (2003) ), WSD senses are thus either words or phrases, as learned in the SMT phrasal translation lexicon. Those \"sense\" candidates are very different from those typically used even in dedicated WSD tasks, even in the multilingual Senseval tasks. Each candidate is a phrase that is not necessarily a syntactic noun or verb phrase as in manually compiled dictionaries. It is quite possible that distinct \"senses\" in our WSD for SMT system could be considered synonyms in a traditional WSD framework, especially in monolingual WSD.",
                "cite_spans": [
                    {
                        "start": 51,
                        "end": 70,
                        "text": "Koehn et al. (2003)",
                        "ref_id": "BIBREF21"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "WSD uses the same sense definitions as the SMT system",
                "sec_num": "4.2"
            },
            {
                "text": "In addition to the consistency requirements for integration, this requirement is also motivated by empirical studies, which show that predefined translations derived from sense distinctions defined in monolingual ontologies do not match translation distinction made by human translators .",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "WSD uses the same sense definitions as the SMT system",
                "sec_num": "4.2"
            },
            {
                "text": "WSD training does not require any other resources than SMT training, nor any manual sense annotation. We employ supervised WSD systems, since Senseval results have amply demonstrated that supervised models significantly outperform unsupervised approaches (see for instance the English lexical sample tasks results described by ).",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "WSD uses the same training data as the SMT system",
                "sec_num": "4.3"
            },
            {
                "text": "Training examples are annotated using the phrase alignments learned during SMT training. Every in-put language phrase is sense-tagged with its aligned output language phrase in the parallel corpus. The phrase alignment method used to extract the WSD training data therefore depends on the one used by the SMT system. This presents the advantage of training WSD and SMT models on exactly the same data, thus eliminating domain mismatches between Senseval data and parallel corpora. But most importantly, this allows WSD training data to be generated entirely automatically, since the parallel corpus is automatically phrase-aligned in order to learn the SMT phrase bilexicon.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "WSD uses the same training data as the SMT system",
                "sec_num": "4.3"
            },
            {
                "text": "The word sense disambiguation subsystem is modeled after the best performing WSD system in the Chinese lexical sample task at Senseval-3 .",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "The features employed are typical of WSD and are therefore far richer than those used in most SMT systems. The feature set consists of positionsensitive, syntactic, and local collocational features, since these features yielded the best results when combined in a na\u00efve Bayes model on several Senseval-2 lexical sample tasks (Yarowsky and Florian, 2002) . These features scale easily to the bigger vocabulary and sense candidates to be considered in a SMT task.",
                "cite_spans": [
                    {
                        "start": 325,
                        "end": 353,
                        "text": "(Yarowsky and Florian, 2002)",
                        "ref_id": "BIBREF44"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "The Senseval system consists of an ensemble of four combined WSD models:",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "The first model is a na\u00efve Bayes model, since Yarowsky and Florian (2002) found this model to be the most accurate classifier in a comparative study on a subset of Senseval-2 English lexical sample data.",
                "cite_spans": [
                    {
                        "start": 46,
                        "end": 73,
                        "text": "Yarowsky and Florian (2002)",
                        "ref_id": "BIBREF44"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "The second model is a maximum entropy model (Jaynes, 1978) , since Klein and Manning (Klein and Manning, 2002) found that this model yielded higher accuracy than na\u00efve Bayes in a subsequent comparison of WSD performance.",
                "cite_spans": [
                    {
                        "start": 44,
                        "end": 58,
                        "text": "(Jaynes, 1978)",
                        "ref_id": "BIBREF17"
                    },
                    {
                        "start": 85,
                        "end": 110,
                        "text": "(Klein and Manning, 2002)",
                        "ref_id": "BIBREF20"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "The third model is a boosting model (Freund and Schapire, 1997) , since boosting has consistently turned in very competitive scores on related tasks such as named entity classification. We also use the Adaboost.MH algorithm.",
                "cite_spans": [
                    {
                        "start": 36,
                        "end": 63,
                        "text": "(Freund and Schapire, 1997)",
                        "ref_id": "BIBREF14"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "The fourth model is a Kernel PCA-based model . Kernel Principal Component Analysis or KPCA is a nonlinear kernel method for extracting nonlinear principal components from vector sets where, conceptually, the n-dimensional input vectors are nonlinearly mapped from their original space R n to a high-dimensional feature space F where linear PCA is performed, yielding a transform by which the input vectors can be mapped nonlinearly to a new set of vectors (Sch\u00f6lkopf et al., 1998) . WSD can be performed by a Nearest Neighbor Classifier in the high-dimensional KPCA feature space.",
                "cite_spans": [
                    {
                        "start": 456,
                        "end": 480,
                        "text": "(Sch\u00f6lkopf et al., 1998)",
                        "ref_id": "BIBREF32"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "All these classifiers have the ability to handle large numbers of sparse features, many of which may be irrelevant. Moreover, the maximum entropy and boosting models are known to be well suited to handling features that are highly interdependent.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "The WSD system",
                "sec_num": "4.4"
            },
            {
                "text": "It is non-trivial to incorporate WSD into an existing phrase-based architecture such as Pharaoh (Koehn, 2004) , since the decoder is not set up to easily accept multiple translation probabilities that are dynamically computed in context-sensitive fashion. For every phrase in a given SMT input sentence, the WSD probabilities can be used as additional feature in a loglinear translation model, in combination with typical context-independent SMT bilexicon probabilities.",
                "cite_spans": [
                    {
                        "start": 96,
                        "end": 109,
                        "text": "(Koehn, 2004)",
                        "ref_id": "BIBREF22"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Integrating WSD predictions in phrase-based SMT architectures",
                "sec_num": "4.5"
            },
            {
                "text": "We overcome this obstacle by devising a calling architecture that reinitializes the decoder with dynamically generated lexicons on a per-sentence basis.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Integrating WSD predictions in phrase-based SMT architectures",
                "sec_num": "4.5"
            },
            {
                "text": "Unlike a n-best reranking approach, which is limited by the lexical choices made by the decoder using only the baseline context-independent translation probabilities, our method allows the system to make full use of WSD information for all competing phrases at all decoding stages.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Integrating WSD predictions in phrase-based SMT architectures",
                "sec_num": "4.5"
            },
            {
                "text": "The evaluation is conducted on two standard Chinese to English translation tasks. We follow standard machine translation evaluation procedure using automatic evaluation metrics. Since our goal is to evaluate translation quality, we use standard MT evaluation methodology and do not evaluate the accuracy of the WSD model independently. Preliminary experiments are conducted using training and evaluation data drawn from the multilingual BTEC corpus, which contains sentences used in conversations in the travel domain, and their translations in several languages. A subset of this data was made available for the IWSLT06 evaluation campaign (Paul, 2006) ; the training set consists of 40000 sentence pairs, and each test set contains around 500 sentences. We used only the pure text data, and not the speech transcriptions, so that speech-specific issues would not interfere with our primary goal of understanding the effect of integrating WSD in a fullscale phrase-based model. A larger scale evaluation is conducted on the standard NIST Chinese-English test set (MT-04), which contains 1788 sentences drawn from newswire corpora, and therefore of a much wider domain than the IWSLT data set. The training set consists of about 1 million sentence pairs in the news domain.",
                "cite_spans": [
                    {
                        "start": 641,
                        "end": 653,
                        "text": "(Paul, 2006)",
                        "ref_id": "BIBREF30"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Experimental setup",
                "sec_num": "5"
            },
            {
                "text": "Basic preprocessing was applied to the corpus. The English side was simply tokenized and casenormalized. The Chinese side was word segmented using the LDC segmenter.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Experimental setup",
                "sec_num": "5"
            },
            {
                "text": "Since our focus is not on a specific SMT architecture, we use the off-the-shelf phrase-based decoder Pharaoh (Koehn, 2004) trained on the IWSLT training set. Pharaoh implements a beam search decoder for phrase-based statistical models, and presents the advantages of being freely available and widely used.",
                "cite_spans": [
                    {
                        "start": 109,
                        "end": 122,
                        "text": "(Koehn, 2004)",
                        "ref_id": "BIBREF22"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Baseline SMT system",
                "sec_num": "5.2"
            },
            {
                "text": "The phrase bilexicon is derived from the intersection of bidirectional IBM Model 4 alignments, obtained with GIZA++ (Och and Ney, 2003) , augmented to improve recall using the grow-diag-final heuristic. The language model is trained on the English side of the corpus using the SRI language modeling toolkit (Stolcke, 2002) .",
                "cite_spans": [
                    {
                        "start": 116,
                        "end": 135,
                        "text": "(Och and Ney, 2003)",
                        "ref_id": "BIBREF27"
                    },
                    {
                        "start": 307,
                        "end": 322,
                        "text": "(Stolcke, 2002)",
                        "ref_id": "BIBREF36"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Baseline SMT system",
                "sec_num": "5.2"
            },
            {
                "text": "The loglinear model weights are learned using Chiang's implementation of the maximum BLEU training algorithm (Och, 2003) , both for the baseline, and the WSD-augmented system. Due to time constraints, this optimization was only conducted on the IWSLT task. The weights used in the WSD-augmented NIST model are based on the best IWSLT model. Given that the two tasks are quite different, we expect further improvements on the WSD-augmented system after running maximum BLEU optimization for the NIST task.",
                "cite_spans": [
                    {
                        "start": 109,
                        "end": 120,
                        "text": "(Och, 2003)",
                        "ref_id": "BIBREF28"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Baseline SMT system",
                "sec_num": "5.2"
            },
            {
                "text": "Using WSD predictions in SMT yields better translation quality on all test sets, as measured by all eight commonly used automatic evaluation metrics. The results are shown in Table 1 for IWSLT and Table 2 for the NIST task. Paired bootstrap resampling shows that the improvements on the NIST test set are statistically significant at the 95% level.",
                "cite_spans": [],
                "ref_spans": [
                    {
                        "start": 175,
                        "end": 182,
                        "text": "Table 1",
                        "ref_id": "TABREF0"
                    }
                ],
                "eq_spans": [],
                "section": "Results and discussion",
                "sec_num": "6"
            },
            {
                "text": "Remarkably, integrating WSD predictions helps all the very different metrics. In addition to the widely used BLEU (Papineni et al., 2002) and NIST (Doddington, 2002) scores, we also evaluate translation quality with the recently proposed Meteor (Banerjee and Lavie, 2005) and four edit-distance style metrics, Word Error Rate (WER), Positionindependent word Error Rate (PER) (Tillmann et al., 1997) , CDER, which allows block reordering (Leusch et al., 2006) , and Translation Edit Rate (TER) (Snover et al., 2006) . Note that we report Meteor scores computed both with and without using WordNet synonyms to match translation candidates and references, showing that the improvement is not due to context-independent synonym matches at evaluation time.",
                "cite_spans": [
                    {
                        "start": 114,
                        "end": 137,
                        "text": "(Papineni et al., 2002)",
                        "ref_id": "BIBREF29"
                    },
                    {
                        "start": 147,
                        "end": 165,
                        "text": "(Doddington, 2002)",
                        "ref_id": "BIBREF12"
                    },
                    {
                        "start": 245,
                        "end": 271,
                        "text": "(Banerjee and Lavie, 2005)",
                        "ref_id": "BIBREF0"
                    },
                    {
                        "start": 375,
                        "end": 398,
                        "text": "(Tillmann et al., 1997)",
                        "ref_id": "BIBREF37"
                    },
                    {
                        "start": 437,
                        "end": 458,
                        "text": "(Leusch et al., 2006)",
                        "ref_id": "BIBREF23"
                    },
                    {
                        "start": 493,
                        "end": 514,
                        "text": "(Snover et al., 2006)",
                        "ref_id": "BIBREF33"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Results and discussion",
                "sec_num": "6"
            },
            {
                "text": "Comparison of the 1-Best decoder output with and without the WSD feature shows that the sentences differ by one or more token respectively for 25.49%, 30.40% and 29.25% of IWSLT test sets 1, The purpose of my visit to Japan is pray for peace and prosperity. SMT+WSD The purpose of my visit is to pray for peace and prosperity for Japan.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Results and discussion",
                "sec_num": "6"
            },
            {
                "text": ": 2 P ;\u00a8 I\u00f6 f\u00b9 \u00c7\u00d6 \u2020 M@* \" %AE \u00dd\u2030 \u00aa\u00bd SMT In order to prevent terrorist activities Los Angeles, the police have taken unprecedented tight security measures. SMT+WSD In order to prevent terrorist activities Los Angeles, the police to an unprecedented tight security measures.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "2 and 3, and 95.74% of the NIST test set. Tables 3 and 4 show examples of translations drawn from the IWSLT and NIST test sets respectively.",
                "cite_spans": [],
                "ref_spans": [
                    {
                        "start": 42,
                        "end": 56,
                        "text": "Tables 3 and 4",
                        "ref_id": "TABREF2"
                    }
                ],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "A more detailed analysis reveals WSD predictions give better rankings and are more discriminative than baseline translation probabilities, which helps the final translation in three different ways.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "\u2022 The rich context features help rank the correct translation first with WSD while it is ranked lower according to baseline translation probability scores .",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "\u2022 Even when WSD and baseline translation probabilities agree on the top translation candidate, the stronger WSD scores help override wrong language model predictions.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "\u2022 The strong WSD scores for phrases help the decoder pick longer phrase translations, while using baseline translation probabilities often translate those phrases in smaller chunks that include a frequent (and incorrect) translation candidate.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "For instance, the top 4 Chinese sentences in Ta-ble 4, are better translated by the WSD-augmented system because the WSD scores help the decoder to choose longer phrases. In the first example, the phrase \"\u00a1 \u00fb U\" is correctly translated as a whole as \"No\" by the WSD-augmented system, while the baseline translates each word separately yielding an incorrect translation. In the following three examples, the WSD system encourages the decoder to translate the long phrases \"\u00f4 / \u00e4 \u017d\u00fd \u00c5\u00e7\", \"\u017d\u00fd \" \u00baC \u00b6\u00b5\", and \"HB \u00e5, \" OEs Ac\" as single units, while the baseline introduces errors by breaking them down into shorter phrases.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "The last sentence in the table shows an example where the WSD predictions do not help the baseline system. The translation quality is actually much worse, since the verb \"\u00c7 \u00d6\" is incorrectly translated as \"to\", despite the fact that the top candidate predicted by the WSD system alone is the much better translation \"has taken\", but with a relatively low probability of 0.509.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Input",
                "sec_num": null
            },
            {
                "text": "We have shown for the first time that integrating multi-word phrasal WSD models into phrase-based SMT consistently helps on all commonly available automated translation quality evaluation metrics on all three different test sets from the Chinese-English IWSLT06 text translation task, and yields statistically significant gains on the larger NIST Chinese-English task. It is important to note that the WSD models never hurt translation quality, and always yield individual gains of a level that makes their integration always worthwile.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Conclusion",
                "sec_num": "7"
            },
            {
                "text": "We have proposed to consistently integrate WSD models both during training, where sense definitions and sense-annotated data are automatically extracted from the word-aligned parallel corpora from SMT training, and during testing, where the phrasal WSD probabilities are used by the SMT system just like all the other lexical choice features.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Conclusion",
                "sec_num": "7"
            },
            {
                "text": "Context features are derived from state-of-the-art WSD models, and the evaluation is conducted on the actual translation task, rather than intermediate tasks such as word alignment.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Conclusion",
                "sec_num": "7"
            },
            {
                "text": "It is to be emphasized that this approach does not merely consist of adding a source sentence feature in the log linear model for translation. On the contrary, it remains a real WSD task, defined just as in the Senseval Multilingual Lexical Sample tasks (e.g., ). Our model makes use of typical WSD features that are almost never used in SMT systems, and requires a dynamically created translation lexicon on a per-sentence basis.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Conclusion",
                "sec_num": "7"
            },
            {
                "text": "To our knowledge this constitues the first attempt at fully integrating state-of-the-art WSD with conventional phrase-based SMT. Unlike previous approaches, the WSD targets are not only single words, but multi-word phrases, just as in the SMT system. This means that WSD senses are unusually predicted not only for a limited set of single words or very short phrases, but for all phrases of arbitrarily length that are in the SMT translation lexicon. The single word approach, as we reported in Carpuat et al. (2006) , improved BLEU and NIST scores for phrase-based SMT, but subsequent detailed empirical studies we have performed since then suggest that single word WSD approaches are less successful when evaluated under all other MT metrics (Carpuat and Wu, 2007) . Thus, fully phrasal WSD predictions for longer phrases, as reported in this paper, are particularly important to improve translation quality.",
                "cite_spans": [
                    {
                        "start": 495,
                        "end": 516,
                        "text": "Carpuat et al. (2006)",
                        "ref_id": "BIBREF7"
                    },
                    {
                        "start": 744,
                        "end": 766,
                        "text": "(Carpuat and Wu, 2007)",
                        "ref_id": "BIBREF5"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Conclusion",
                "sec_num": "7"
            },
            {
                "text": "The results reported in this paper cast new light on the WSD vs. SMT debate, suggesting that a close integration of WSD and SMT decisions should be incorporated in a SMT model that successfully uses WSD predictions. Our objective here is to demonstrate that this technique works for the widest possible class of models, so we have chosen as the baseline the most widely used phrase-based SMT model. Our positive results suggest that our experiments could be tried on other current statistical MT models, especially the growing family of treestructured SMT models employing stochastic transduction grammars of various sorts (Wu and Chiang, 2007) . For instance, incorporating WSD predictions into an MT decoder based on inversion transduction grammars (Wu, 1997) -such as the Bracketing ITG based models of Wu (1996) , Zens et al. (2004) , or Cherry and Lin (2007) -would present an intriguing comparison with the present work. It would also be interesting to assess whether a more grammatically structured statistical MT model that is less reliant on an n-gram language model, such as the syntactic ITG based \"grammatical channel\" translation model of (Wu and Wong, 1998) , could make more effective use of WSD predictions.",
                "cite_spans": [
                    {
                        "start": 623,
                        "end": 644,
                        "text": "(Wu and Chiang, 2007)",
                        "ref_id": "BIBREF39"
                    },
                    {
                        "start": 751,
                        "end": 761,
                        "text": "(Wu, 1997)",
                        "ref_id": "BIBREF43"
                    },
                    {
                        "start": 806,
                        "end": 815,
                        "text": "Wu (1996)",
                        "ref_id": "BIBREF42"
                    },
                    {
                        "start": 818,
                        "end": 836,
                        "text": "Zens et al. (2004)",
                        "ref_id": "BIBREF45"
                    },
                    {
                        "start": 842,
                        "end": 863,
                        "text": "Cherry and Lin (2007)",
                        "ref_id": "BIBREF8"
                    },
                    {
                        "start": 1152,
                        "end": 1171,
                        "text": "(Wu and Wong, 1998)",
                        "ref_id": "BIBREF40"
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Conclusion",
                "sec_num": "7"
            },
            {
                "text": "RGC6083/99E, RGC6256/00E, and DAG03/04.EG09. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Defense Advanced Research Projects Agency.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "",
                "sec_num": null
            }
        ],
        "back_matter": [],
        "bib_entries": {
            "BIBREF0": {
                "ref_id": "b0",
                "title": "METEOR: An automatic metric for MT evaluation with improved correlation with human judgement",
                "authors": [
                    {
                        "first": "Satanjeev",
                        "middle": [],
                        "last": "Banerjee",
                        "suffix": ""
                    },
                    {
                        "first": "Alon",
                        "middle": [],
                        "last": "Lavie",
                        "suffix": ""
                    }
                ],
                "year": 2005,
                "venue": "Proceedings of Workshop on Intrinsic and Extrinsic Evaluation Measures for MT and/or Summarization at the 43th Annual Meeting of the Association of Computational Linguistics (ACL-2005)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Satanjeev Banerjee and Alon Lavie. METEOR: An automatic metric for MT evaluation with im- proved correlation with human judgement. In Proceedings of Workshop on Intrinsic and Extrin- sic Evaluation Measures for MT and/or Summa- rization at the 43th Annual Meeting of the Associ- ation of Computational Linguistics (ACL-2005), Ann Arbor, Michigan, June 2005.",
                "links": null
            },
            "BIBREF1": {
                "ref_id": "b1",
                "title": "Word-sense disambiguation using statistical methods",
                "authors": [
                    {
                        "first": "Peter",
                        "middle": [],
                        "last": "Brown",
                        "suffix": ""
                    },
                    {
                        "first": "Stephen",
                        "middle": [
                            "Della"
                        ],
                        "last": "Pietra",
                        "suffix": ""
                    },
                    {
                        "first": "Vincent",
                        "middle": [
                            "Della"
                        ],
                        "last": "Pietra",
                        "suffix": ""
                    },
                    {
                        "first": "Robert",
                        "middle": [],
                        "last": "Mercer",
                        "suffix": ""
                    }
                ],
                "year": 1991,
                "venue": "Proceedings of 29th meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "264--270",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Peter Brown, Stephen Della Pietra, Vincent Della Pietra, and Robert Mercer. Word-sense disam- biguation using statistical methods. In Proceed- ings of 29th meeting of the Association for Com- putational Linguistics, pages 264-270, Berkeley, California, 1991.",
                "links": null
            },
            "BIBREF2": {
                "ref_id": "b2",
                "title": "Using WSD techniques for lexical selection in statistical machine translation",
                "authors": [
                    {
                        "first": "Clara",
                        "middle": [],
                        "last": "Cabezas",
                        "suffix": ""
                    },
                    {
                        "first": "Philip",
                        "middle": [],
                        "last": "Resnik",
                        "suffix": ""
                    }
                ],
                "year": 2005,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Clara Cabezas and Philip Resnik. Using WSD tech- niques for lexical selection in statistical machine translation. Technical report, Institute for Ad- vanced Computer Studies, University of Mary- land, 2005.",
                "links": null
            },
            "BIBREF3": {
                "ref_id": "b3",
                "title": "Evaluating the word sense disambiguation performance of statistical machine translation",
                "authors": [
                    {
                        "first": "Marine",
                        "middle": [],
                        "last": "Carpuat",
                        "suffix": ""
                    },
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 2005,
                "venue": "Proceedings of the Second International Joint Conference on Natural Language Processing (IJCNLP)",
                "volume": "",
                "issue": "",
                "pages": "122--127",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Marine Carpuat and Dekai Wu. Evaluating the word sense disambiguation performance of statistical machine translation. In Proceedings of the Second International Joint Conference on Natural Lan- guage Processing (IJCNLP), pages 122-127, Jeju Island, Republic of Korea, 2005.",
                "links": null
            },
            "BIBREF4": {
                "ref_id": "b4",
                "title": "Word sense disambiguation vs. statistical machine translation",
                "authors": [
                    {
                        "first": "Marine",
                        "middle": [],
                        "last": "Carpuat",
                        "suffix": ""
                    },
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 2005,
                "venue": "Proceedings of the annual meeting of the association for computational linguistics (ACL-05)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Marine Carpuat and Dekai Wu. Word sense disam- biguation vs. statistical machine translation. In Proceedings of the annual meeting of the associa- tion for computational linguistics (ACL-05), Ann Arbor, Michigan, 2005.",
                "links": null
            },
            "BIBREF5": {
                "ref_id": "b5",
                "title": "How phrase sense disambiguation outperforms word sense disambiguation for statistical machine translation. Forthcoming",
                "authors": [
                    {
                        "first": "Marine",
                        "middle": [],
                        "last": "Carpuat",
                        "suffix": ""
                    },
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 2007,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Marine Carpuat and Dekai Wu. How phrase sense disambiguation outperforms word sense disambiguation for statistical machine translation. Forthcoming, 2007.",
                "links": null
            },
            "BIBREF6": {
                "ref_id": "b6",
                "title": "Augmenting ensemble classification for word sense disambiguation with a Kernel PCA model",
                "authors": [
                    {
                        "first": "Marine",
                        "middle": [],
                        "last": "Carpuat",
                        "suffix": ""
                    },
                    {
                        "first": "Weifeng",
                        "middle": [],
                        "last": "Su",
                        "suffix": ""
                    },
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "Proceedings of Senseval-3, Third International Workshop on Evaluating Word Sense Disambiguation Systems",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Marine Carpuat, Weifeng Su, and Dekai Wu. Aug- menting ensemble classification for word sense disambiguation with a Kernel PCA model. In Pro- ceedings of Senseval-3, Third International Work- shop on Evaluating Word Sense Disambiguation Systems, Barcelona, July 2004. SIGLEX, Associ- ation for Computational Linguistics.",
                "links": null
            },
            "BIBREF7": {
                "ref_id": "b7",
                "title": "Toward integrating word sense and entity disambiguation into statistical machine translation",
                "authors": [
                    {
                        "first": "Marine",
                        "middle": [],
                        "last": "Carpuat",
                        "suffix": ""
                    },
                    {
                        "first": "Yihai",
                        "middle": [],
                        "last": "Shen",
                        "suffix": ""
                    },
                    {
                        "first": "Xiaofeng",
                        "middle": [],
                        "last": "Yu",
                        "suffix": ""
                    },
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 2006,
                "venue": "Third International Workshop on Spoken Language Translation (IWSLT 2006)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Marine Carpuat, Yihai Shen, Xiaofeng Yu, and Dekai Wu. Toward integrating word sense and en- tity disambiguation into statistical machine trans- lation. In Third International Workshop on Spo- ken Language Translation (IWSLT 2006), Kyoto, November 2006.",
                "links": null
            },
            "BIBREF8": {
                "ref_id": "b8",
                "title": "Inversion Transduction Grammar for joint phrasal translation modeling",
                "authors": [
                    {
                        "first": "Colin",
                        "middle": [],
                        "last": "Cherry",
                        "suffix": ""
                    },
                    {
                        "first": "Dekang",
                        "middle": [],
                        "last": "Lin",
                        "suffix": ""
                    }
                ],
                "year": 2007,
                "venue": "NAACL-HLT 2007 / AMTA Workshop on Syntax and Structure in Statistical Translation (SSST)",
                "volume": "",
                "issue": "",
                "pages": "17--24",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Colin Cherry and Dekang Lin. Inversion Transduc- tion Grammar for joint phrasal translation mod- eling. In Dekai Wu and David Chiang, editors, NAACL-HLT 2007 / AMTA Workshop on Syntax and Structure in Statistical Translation (SSST), pages 17-24, Rochester, NY, April 2007.",
                "links": null
            },
            "BIBREF9": {
                "ref_id": "b9",
                "title": "The Senseval-3 multilingual English-Hindi lexical sample task",
                "authors": [
                    {
                        "first": "Timothy",
                        "middle": [],
                        "last": "Chklovski",
                        "suffix": ""
                    },
                    {
                        "first": "Rada",
                        "middle": [],
                        "last": "Mihalcea",
                        "suffix": ""
                    },
                    {
                        "first": "Ted",
                        "middle": [],
                        "last": "Pedersen",
                        "suffix": ""
                    },
                    {
                        "first": "Amruta",
                        "middle": [],
                        "last": "Purandare",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "Proceedings of Senseval-3, Third International Workshop on Evaluating Word Sense Disambiguation Systems",
                "volume": "",
                "issue": "",
                "pages": "5--8",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Timothy Chklovski, Rada Mihalcea, Ted Pedersen, and Amruta Purandare. The Senseval-3 multilin- gual English-Hindi lexical sample task. In Pro- ceedings of Senseval-3, Third International Work- shop on Evaluating Word Sense Disambiguation Systems, pages 5-8, Barcelona, Spain, July 2004. SIGLEX, Association for Computational Linguis- tics.",
                "links": null
            },
            "BIBREF10": {
                "ref_id": "b10",
                "title": "Word sense disambiguation using a second language monolingual corpus",
                "authors": [
                    {
                        "first": "Ido",
                        "middle": [],
                        "last": "Dagan",
                        "suffix": ""
                    },
                    {
                        "first": "Alon",
                        "middle": [],
                        "last": "Itai",
                        "suffix": ""
                    }
                ],
                "year": 1994,
                "venue": "Computational Linguistics",
                "volume": "20",
                "issue": "4",
                "pages": "563--596",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Ido Dagan and Alon Itai. Word sense disambigua- tion using a second language monolingual corpus. Computational Linguistics, 20(4):563-596, 1994.",
                "links": null
            },
            "BIBREF11": {
                "ref_id": "b11",
                "title": "Relieving the data acquisition bottleneck in word sense disambiguation",
                "authors": [
                    {
                        "first": "Mona",
                        "middle": [],
                        "last": "Diab",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Mona Diab. Relieving the data acquisition bottle- neck in word sense disambiguation. In Proceed- ings of the 42nd Annual Meeting of the Associa- tion for Computational Linguistics, 2004.",
                "links": null
            },
            "BIBREF12": {
                "ref_id": "b12",
                "title": "Automatic evaluation of machine translation quality using n-gram cooccurrence statistics",
                "authors": [
                    {
                        "first": "George",
                        "middle": [],
                        "last": "Doddington",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "Proceedings of the Human Language Technology conference (HLT-2002)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "George Doddington. Automatic evaluation of machine translation quality using n-gram co- occurrence statistics. In Proceedings of the Human Language Technology conference (HLT- 2002), San Diego, CA, 2002.",
                "links": null
            },
            "BIBREF13": {
                "ref_id": "b13",
                "title": "Knowledge description: what, how and who?",
                "authors": [
                    {
                        "first": "Zhendong",
                        "middle": [],
                        "last": "Dong",
                        "suffix": ""
                    }
                ],
                "year": 1998,
                "venue": "Proceedings of International Symposium on Electronic Dictionary",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Zhendong Dong. Knowledge description: what, how and who? In Proceedings of Inter- national Symposium on Electronic Dictionary, Tokyo, Japan, 1998.",
                "links": null
            },
            "BIBREF14": {
                "ref_id": "b14",
                "title": "A decisiontheoretic generalization of on-line learning and an application to boosting",
                "authors": [
                    {
                        "first": "Yoram",
                        "middle": [],
                        "last": "Freund",
                        "suffix": ""
                    },
                    {
                        "first": "Robert",
                        "middle": [
                            "E"
                        ],
                        "last": "Schapire",
                        "suffix": ""
                    }
                ],
                "year": 1997,
                "venue": "In Journal of Computer and System Sciences",
                "volume": "55",
                "issue": "1",
                "pages": "119--139",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Yoram Freund and Robert E. Schapire. A decision- theoretic generalization of on-line learning and an application to boosting. In Journal of Com- puter and System Sciences, 55(1), pages 119-139, 1997.",
                "links": null
            },
            "BIBREF15": {
                "ref_id": "b15",
                "title": "Refined lexicon models for statistical machine translation using a maximum entropy approach",
                "authors": [
                    {
                        "first": "Ismael",
                        "middle": [],
                        "last": "Garcia-Varea",
                        "suffix": ""
                    },
                    {
                        "first": "Franz",
                        "middle": [],
                        "last": "Och",
                        "suffix": ""
                    },
                    {
                        "first": "Hermann",
                        "middle": [],
                        "last": "Ney",
                        "suffix": ""
                    },
                    {
                        "first": "Francisco",
                        "middle": [],
                        "last": "Casacuberta",
                        "suffix": ""
                    }
                ],
                "year": 2001,
                "venue": "Proceedings of the 39th annual meeting of the association for computational linguistics (ACL-01)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Ismael Garcia-Varea, Franz Och, Hermann Ney, and Francisco Casacuberta. Refined lexicon models for statistical machine translation using a maxi- mum entropy approach. In Proceedings of the 39th annual meeting of the association for compu- tational linguistics (ACL-01), Toulouse, France, 2001.",
                "links": null
            },
            "BIBREF16": {
                "ref_id": "b16",
                "title": "Efficient integration of maximum entropy lexicon models within the training of statistical alignment models",
                "authors": [
                    {
                        "first": "Ismael",
                        "middle": [],
                        "last": "Garcia-Varea",
                        "suffix": ""
                    },
                    {
                        "first": "Franz",
                        "middle": [],
                        "last": "Och",
                        "suffix": ""
                    },
                    {
                        "first": "Hermann",
                        "middle": [],
                        "last": "Ney",
                        "suffix": ""
                    },
                    {
                        "first": "Francisco",
                        "middle": [],
                        "last": "Casacuberta",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "Proceedings of AMTA-2002",
                "volume": "",
                "issue": "",
                "pages": "54--63",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Ismael Garcia-Varea, Franz Och, Hermann Ney, and Francisco Casacuberta. Efficient integration of maximum entropy lexicon models within the training of statistical alignment models. In Pro- ceedings of AMTA-2002, pages 54-63, Tiburon, California, October 2002.",
                "links": null
            },
            "BIBREF17": {
                "ref_id": "b17",
                "title": "Where do we Stand on Maximum Entropy?",
                "authors": [
                    {
                        "first": "E",
                        "middle": [
                            "T"
                        ],
                        "last": "Jaynes",
                        "suffix": ""
                    }
                ],
                "year": 1978,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "E.T. Jaynes. Where do we Stand on Maximum En- tropy? MIT Press, Cambridge MA, 1978.",
                "links": null
            },
            "BIBREF18": {
                "ref_id": "b18",
                "title": "Framework and results for English Senseval",
                "authors": [
                    {
                        "first": "Adam",
                        "middle": [],
                        "last": "Kilgarriff",
                        "suffix": ""
                    },
                    {
                        "first": "Joseph",
                        "middle": [],
                        "last": "Rosenzweig",
                        "suffix": ""
                    }
                ],
                "year": 1999,
                "venue": "Computers and the Humanities",
                "volume": "34",
                "issue": "1",
                "pages": "15--48",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Adam Kilgarriff and Joseph Rosenzweig. Frame- work and results for English Senseval. Computers and the Humanities, 34(1):15-48, 1999. Special issue on SENSEVAL.",
                "links": null
            },
            "BIBREF19": {
                "ref_id": "b19",
                "title": "English lexical sample task description",
                "authors": [
                    {
                        "first": "Adam",
                        "middle": [],
                        "last": "Kilgarriff",
                        "suffix": ""
                    }
                ],
                "year": 2001,
                "venue": "Second International Workshop on Evaluating Word Sense Disambiguation Systems",
                "volume": "",
                "issue": "",
                "pages": "17--20",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Adam Kilgarriff. English lexical sample task de- scription. In Proceedings of Senseval-2, Sec- ond International Workshop on Evaluating Word Sense Disambiguation Systems, pages 17-20, Toulouse, France, July 2001. SIGLEX, Associa- tion for Computational Linguistics.",
                "links": null
            },
            "BIBREF20": {
                "ref_id": "b20",
                "title": "Conditional structure versus conditional estimation in NLP models",
                "authors": [
                    {
                        "first": "Dan",
                        "middle": [],
                        "last": "Klein",
                        "suffix": ""
                    },
                    {
                        "first": "Christopher",
                        "middle": [
                            "D"
                        ],
                        "last": "Manning",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "Proceedings of EMNLP-2002, Conference on Empirical Methods in Natural Language Processing",
                "volume": "",
                "issue": "",
                "pages": "9--16",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Dan Klein and Christopher D. Manning. Conditional structure versus conditional estimation in NLP models. In Proceedings of EMNLP-2002, Confer- ence on Empirical Methods in Natural Language Processing, pages 9-16, Philadelphia, July 2002. SIGDAT, Association for Computational Linguis- tics.",
                "links": null
            },
            "BIBREF21": {
                "ref_id": "b21",
                "title": "Statistical phrase-based translation",
                "authors": [
                    {
                        "first": "Philipp",
                        "middle": [],
                        "last": "Koehn",
                        "suffix": ""
                    },
                    {
                        "first": "Franz",
                        "middle": [],
                        "last": "Och",
                        "suffix": ""
                    },
                    {
                        "first": "Daniel",
                        "middle": [],
                        "last": "Marcu",
                        "suffix": ""
                    }
                ],
                "year": 2003,
                "venue": "Proceedings of HLT/NAACL-2003",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Philipp Koehn, Franz Och, and Daniel Marcu. Sta- tistical phrase-based translation. In Proceedings of HLT/NAACL-2003, Edmonton, Canada, May 2003.",
                "links": null
            },
            "BIBREF22": {
                "ref_id": "b22",
                "title": "Pharaoh: a beam search decoder for phrase-based statistical machine translation models",
                "authors": [
                    {
                        "first": "Philipp",
                        "middle": [],
                        "last": "Koehn",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "6th Conference of the Association for Machine Translation in the Americas (AMTA)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Philipp Koehn. Pharaoh: a beam search decoder for phrase-based statistical machine translation mod- els. In 6th Conference of the Association for Ma- chine Translation in the Americas (AMTA), Wash- ington, DC, September 2004.",
                "links": null
            },
            "BIBREF23": {
                "ref_id": "b23",
                "title": "Efficient MT evaluation using block movements",
                "authors": [
                    {
                        "first": "Gregor",
                        "middle": [],
                        "last": "Leusch",
                        "suffix": ""
                    },
                    {
                        "first": "Nicola",
                        "middle": [],
                        "last": "Ueffing",
                        "suffix": ""
                    },
                    {
                        "first": "Hermann",
                        "middle": [],
                        "last": "Ney",
                        "suffix": ""
                    }
                ],
                "year": 2006,
                "venue": "Proceedings of EACL-2006 (11th Conference of the European Chapter of the Association for Computational Linguistics)",
                "volume": "",
                "issue": "",
                "pages": "241--248",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Gregor Leusch, Nicola Ueffing, and Hermann Ney. Efficient MT evaluation using block movements. In Proceedings of EACL-2006 (11th Confer- ence of the European Chapter of the Association for Computational Linguistics), pages 241-248, Trento, Italy, April 2006.",
                "links": null
            },
            "BIBREF24": {
                "ref_id": "b24",
                "title": "Word translation disambiguation using bilingual bootstrapping",
                "authors": [
                    {
                        "first": "Cong",
                        "middle": [],
                        "last": "Li",
                        "suffix": ""
                    },
                    {
                        "first": "Hang",
                        "middle": [],
                        "last": "Li",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "343--351",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Cong Li and Hang Li. Word translation disambigua- tion using bilingual bootstrapping. In Proceed- ings of the 40th Annual Meeting of the Associ- ation for Computational Linguistics, pages 343- 351, 2002.",
                "links": null
            },
            "BIBREF25": {
                "ref_id": "b25",
                "title": "The Senseval-3 English lexical sample task",
                "authors": [
                    {
                        "first": "Rada",
                        "middle": [],
                        "last": "Mihalcea",
                        "suffix": ""
                    },
                    {
                        "first": "Timothy",
                        "middle": [],
                        "last": "Chklovski",
                        "suffix": ""
                    },
                    {
                        "first": "Adam",
                        "middle": [],
                        "last": "Killgariff",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "Proceedings of Senseval-3, Third International Workshop on Evaluating Word Sense Disambiguation Systems",
                "volume": "",
                "issue": "",
                "pages": "25--28",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Rada Mihalcea, Timothy Chklovski, and Adam Kill- gariff. The Senseval-3 English lexical sample task. In Proceedings of Senseval-3, Third Interna- tional Workshop on Evaluating Word Sense Dis- ambiguation Systems, pages 25-28, Barcelona, Spain, July 2004. SIGLEX, Association for Com- putational Linguistics.",
                "links": null
            },
            "BIBREF26": {
                "ref_id": "b26",
                "title": "Exploiting parallel texts for word sense disambiguation: An empirical study",
                "authors": [
                    {
                        "first": "Bin",
                        "middle": [],
                        "last": "Hwee Tou Ng",
                        "suffix": ""
                    },
                    {
                        "first": "Yee Seng",
                        "middle": [],
                        "last": "Wang",
                        "suffix": ""
                    },
                    {
                        "first": "",
                        "middle": [],
                        "last": "Chan",
                        "suffix": ""
                    }
                ],
                "year": 2003,
                "venue": "Proceedings of ACL-03",
                "volume": "",
                "issue": "",
                "pages": "455--462",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Hwee Tou Ng, Bin Wang, and Yee Seng Chan. Ex- ploiting parallel texts for word sense disambigua- tion: An empirical study. In Proceedings of ACL- 03, Sapporo, Japan, pages 455-462, 2003.",
                "links": null
            },
            "BIBREF27": {
                "ref_id": "b27",
                "title": "A systematic comparison of various statistical alignment models",
                "authors": [
                    {
                        "first": "Josef",
                        "middle": [],
                        "last": "Franz",
                        "suffix": ""
                    },
                    {
                        "first": "Hermann",
                        "middle": [],
                        "last": "Och",
                        "suffix": ""
                    },
                    {
                        "first": "",
                        "middle": [],
                        "last": "Ney",
                        "suffix": ""
                    }
                ],
                "year": 2003,
                "venue": "Computational Linguistics",
                "volume": "29",
                "issue": "1",
                "pages": "19--52",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Franz Josef Och and Hermann Ney. A system- atic comparison of various statistical alignment models. Computational Linguistics, 29(1):19-52, 2003.",
                "links": null
            },
            "BIBREF28": {
                "ref_id": "b28",
                "title": "Minimum error rate training in statistical machine translation",
                "authors": [
                    {
                        "first": "Franz Josef",
                        "middle": [],
                        "last": "Och",
                        "suffix": ""
                    }
                ],
                "year": 2003,
                "venue": "Proceedings of the 41st Annual Meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "160--167",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Franz Josef Och. Minimum error rate training in sta- tistical machine translation. In Proceedings of the 41st Annual Meeting of the Association for Com- putational Linguistics, pages 160-167, 2003.",
                "links": null
            },
            "BIBREF29": {
                "ref_id": "b29",
                "title": "BLEU: a method for automatic evaluation of machine translation",
                "authors": [
                    {
                        "first": "Kishore",
                        "middle": [],
                        "last": "Papineni",
                        "suffix": ""
                    },
                    {
                        "first": "Salim",
                        "middle": [],
                        "last": "Roukos",
                        "suffix": ""
                    },
                    {
                        "first": "Todd",
                        "middle": [],
                        "last": "Ward",
                        "suffix": ""
                    },
                    {
                        "first": "Wei-Jing",
                        "middle": [],
                        "last": "Zhu",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. BLEU: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, 2002.",
                "links": null
            },
            "BIBREF30": {
                "ref_id": "b30",
                "title": "Overview of the IWSLT06 evaluation campaign",
                "authors": [
                    {
                        "first": "Michael",
                        "middle": [],
                        "last": "Paul",
                        "suffix": ""
                    }
                ],
                "year": 2006,
                "venue": "Third International Workshop on Spoken Language Translation (IWSLT 2006)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Michael Paul. Overview of the IWSLT06 evalua- tion campaign. In Third International Workshop on Spoken Language Translation (IWSLT 2006), Kyoto, November 2006.",
                "links": null
            },
            "BIBREF31": {
                "ref_id": "b31",
                "title": "Distinguising systems and distinguishing senses: New evaluation methods for word sense disambiguation",
                "authors": [
                    {
                        "first": "Philip",
                        "middle": [],
                        "last": "Resnik",
                        "suffix": ""
                    },
                    {
                        "first": "David",
                        "middle": [],
                        "last": "Yarowsky",
                        "suffix": ""
                    }
                ],
                "year": 1999,
                "venue": "Natural Language Engineering",
                "volume": "5",
                "issue": "2",
                "pages": "113--133",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Philip Resnik and David Yarowsky. Distinguising systems and distinguishing senses: New evalua- tion methods for word sense disambiguation. Nat- ural Language Engineering, 5(2):113-133, 1999.",
                "links": null
            },
            "BIBREF32": {
                "ref_id": "b32",
                "title": "Nonlinear component analysis as a kernel eigenvalue problem",
                "authors": [
                    {
                        "first": "Bernhard",
                        "middle": [],
                        "last": "Sch\u00f6lkopf",
                        "suffix": ""
                    },
                    {
                        "first": "Alexander",
                        "middle": [],
                        "last": "Smola",
                        "suffix": ""
                    },
                    {
                        "first": "Klaus-Rober",
                        "middle": [],
                        "last": "M\u00fcller",
                        "suffix": ""
                    }
                ],
                "year": 1998,
                "venue": "Neural Computation",
                "volume": "10",
                "issue": "5",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Bernhard Sch\u00f6lkopf, Alexander Smola, and Klaus- Rober M\u00fcller. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation, 10(5), 1998.",
                "links": null
            },
            "BIBREF33": {
                "ref_id": "b33",
                "title": "A study of translation edit rate with targeted human annotation",
                "authors": [
                    {
                        "first": "Matthew",
                        "middle": [],
                        "last": "Snover",
                        "suffix": ""
                    },
                    {
                        "first": "Bonnie",
                        "middle": [],
                        "last": "Dorr",
                        "suffix": ""
                    },
                    {
                        "first": "Richard",
                        "middle": [],
                        "last": "Schwartz",
                        "suffix": ""
                    },
                    {
                        "first": "Linnea",
                        "middle": [],
                        "last": "Micciulla",
                        "suffix": ""
                    },
                    {
                        "first": "John",
                        "middle": [],
                        "last": "Makhoul",
                        "suffix": ""
                    }
                ],
                "year": 2006,
                "venue": "Proceedings of AMTA",
                "volume": "",
                "issue": "",
                "pages": "223--231",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Matthew Snover, Bonnie Dorr, Richard Schwartz, Linnea Micciulla, and John Makhoul. A study of translation edit rate with targeted human an- notation. In Proceedings of AMTA, pages 223- 231, Boston, MA, 2006. Association for Machine Translation in the Americas.",
                "links": null
            },
            "BIBREF34": {
                "ref_id": "b34",
                "title": "Multilingual versus monolingual WSD",
                "authors": [
                    {
                        "first": "Lucia",
                        "middle": [],
                        "last": "Specia",
                        "suffix": ""
                    },
                    {
                        "first": "Maria",
                        "middle": [],
                        "last": "Das Gracas Volpe",
                        "suffix": ""
                    },
                    {
                        "first": "Gabriela",
                        "middle": [],
                        "last": "Nunes",
                        "suffix": ""
                    },
                    {
                        "first": "Mark",
                        "middle": [],
                        "last": "Castelo Branco Ribeiro",
                        "suffix": ""
                    },
                    {
                        "first": "",
                        "middle": [],
                        "last": "Stevenson",
                        "suffix": ""
                    }
                ],
                "year": 2006,
                "venue": "EACL-2006 Workshop on Making Sense of Sense: Bringing Psycholinguistics and Computational Linguistics Together",
                "volume": "",
                "issue": "",
                "pages": "33--40",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Lucia Specia, Maria das Gracas Volpe Nunes, Gabriela Castelo Branco Ribeiro, and Mark Stevenson. Multilingual versus monolingual WSD. In EACL-2006 Workshop on Making Sense of Sense: Bringing Psycholinguistics and Computational Linguistics Together, pages 33- 40, Trento, Italy, April 2006.",
                "links": null
            },
            "BIBREF35": {
                "ref_id": "b35",
                "title": "A hybrid relational approach for WSD-first results",
                "authors": [
                    {
                        "first": "Lucia",
                        "middle": [],
                        "last": "Specia",
                        "suffix": ""
                    }
                ],
                "year": 2006,
                "venue": "Proceedings of the COL-ING/ACL 06 Student Research Workshop",
                "volume": "",
                "issue": "",
                "pages": "55--60",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Lucia Specia. A hybrid relational approach for WSD-first results. In Proceedings of the COL- ING/ACL 06 Student Research Workshop, pages 55-60, Sydney, July 2006. ACL.",
                "links": null
            },
            "BIBREF36": {
                "ref_id": "b36",
                "title": "SRILM-an extensible language modeling toolkit",
                "authors": [
                    {
                        "first": "Andreas",
                        "middle": [],
                        "last": "Stolcke",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "International Conference on Spoken Language Processing",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Andreas Stolcke. SRILM-an extensible language modeling toolkit. In International Conference on Spoken Language Processing, Denver, Colorado, September 2002.",
                "links": null
            },
            "BIBREF37": {
                "ref_id": "b37",
                "title": "Accelerated DPbased search for statistical translation",
                "authors": [
                    {
                        "first": "Christoph",
                        "middle": [],
                        "last": "Tillmann",
                        "suffix": ""
                    },
                    {
                        "first": "Stefan",
                        "middle": [],
                        "last": "Vogel",
                        "suffix": ""
                    },
                    {
                        "first": "Hermann",
                        "middle": [],
                        "last": "Ney",
                        "suffix": ""
                    },
                    {
                        "first": "A",
                        "middle": [],
                        "last": "Zubiaga",
                        "suffix": ""
                    },
                    {
                        "first": "H",
                        "middle": [],
                        "last": "Sawaf",
                        "suffix": ""
                    }
                ],
                "year": 1997,
                "venue": "Proceedings of Eurospeech'97",
                "volume": "",
                "issue": "",
                "pages": "2667--2670",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Christoph Tillmann, Stefan Vogel, Hermann Ney, A. Zubiaga, and H. Sawaf. Accelerated DP- based search for statistical translation. In Pro- ceedings of Eurospeech'97, pages 2667-2670, Rhodes, Greece, 1997.",
                "links": null
            },
            "BIBREF38": {
                "ref_id": "b38",
                "title": "Word-sense disambiguation for machine translation",
                "authors": [
                    {
                        "first": "David",
                        "middle": [],
                        "last": "Vickrey",
                        "suffix": ""
                    },
                    {
                        "first": "Luke",
                        "middle": [],
                        "last": "Biewald",
                        "suffix": ""
                    },
                    {
                        "first": "Marc",
                        "middle": [],
                        "last": "Teyssier",
                        "suffix": ""
                    },
                    {
                        "first": "Daphne",
                        "middle": [],
                        "last": "Koller",
                        "suffix": ""
                    }
                ],
                "year": 2005,
                "venue": "Joint Human Language Technology conference and Conference on Empirical Methods in Natural Language Processing",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "David Vickrey, Luke Biewald, Marc Teyssier, and Daphne Koller. Word-sense disambiguation for machine translation. In Joint Human Language Technology conference and Conference on Em- pirical Methods in Natural Language Processing (HLT/EMNLP 2005), Vancouver, 2005.",
                "links": null
            },
            "BIBREF39": {
                "ref_id": "b39",
                "title": "NAACL-HLT 2007 / AMTA Workshop on Syntax and Structure in Statistical Translation (SSST)",
                "authors": [
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    },
                    {
                        "first": "David",
                        "middle": [],
                        "last": "Chiang",
                        "suffix": ""
                    }
                ],
                "year": 2007,
                "venue": "",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Dekai Wu and David Chiang, editors. NAACL-HLT 2007 / AMTA Workshop on Syntax and Structure in Statistical Translation (SSST). Association for Computational Linguistics, Rochester, NY, USA, April 2007.",
                "links": null
            },
            "BIBREF40": {
                "ref_id": "b40",
                "title": "Machine translation with a stochastic grammatical channel",
                "authors": [
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    },
                    {
                        "first": "Hongsing",
                        "middle": [],
                        "last": "Wong",
                        "suffix": ""
                    }
                ],
                "year": 1998,
                "venue": "Proceedings of COLING-ACL'98",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Dekai Wu and Hongsing Wong. Machine translation with a stochastic grammatical channel. In Pro- ceedings of COLING-ACL'98, Montreal,Canada, August 1998.",
                "links": null
            },
            "BIBREF41": {
                "ref_id": "b41",
                "title": "A Kernel PCA method for superior word sense disambiguation",
                "authors": [
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    },
                    {
                        "first": "Weifeng",
                        "middle": [],
                        "last": "Su",
                        "suffix": ""
                    },
                    {
                        "first": "Marine",
                        "middle": [],
                        "last": "Carpuat",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Dekai Wu, Weifeng Su, and Marine Carpuat. A Kernel PCA method for superior word sense dis- ambiguation. In Proceedings of the 42nd An- nual Meeting of the Association for Computa- tional Linguistics, Barcelona, Spain, July 2004.",
                "links": null
            },
            "BIBREF42": {
                "ref_id": "b42",
                "title": "A polynomial-time algorithm for statistical machine translation",
                "authors": [
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 1996,
                "venue": "Proceedings of 34th Annual Meeting of the Association for Computational Linguistics",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Dekai Wu. A polynomial-time algorithm for statis- tical machine translation. In Proceedings of 34th Annual Meeting of the Association for Computa- tional Linguistics, Santa Cruz, California, June 1996.",
                "links": null
            },
            "BIBREF43": {
                "ref_id": "b43",
                "title": "Stochastic inversion transduction grammars and bilingual parsing of parallel corpora",
                "authors": [
                    {
                        "first": "Dekai",
                        "middle": [],
                        "last": "Wu",
                        "suffix": ""
                    }
                ],
                "year": 1997,
                "venue": "Computational Linguistics",
                "volume": "23",
                "issue": "3",
                "pages": "377--404",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Dekai Wu. Stochastic inversion transduction gram- mars and bilingual parsing of parallel corpora. Computational Linguistics, 23(3):377-404, 1997.",
                "links": null
            },
            "BIBREF44": {
                "ref_id": "b44",
                "title": "Evaluating sense disambiguation across diverse parameter spaces",
                "authors": [
                    {
                        "first": "David",
                        "middle": [],
                        "last": "Yarowsky",
                        "suffix": ""
                    },
                    {
                        "first": "Radu",
                        "middle": [],
                        "last": "Florian",
                        "suffix": ""
                    }
                ],
                "year": 2002,
                "venue": "Natural Language Engineering",
                "volume": "8",
                "issue": "4",
                "pages": "293--310",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "David Yarowsky and Radu Florian. Evaluating sense disambiguation across diverse parameter spaces. Natural Language Engineering, 8(4):293-310, 2002.",
                "links": null
            },
            "BIBREF45": {
                "ref_id": "b45",
                "title": "Reordering constraints for phrase-based statistical machine translation",
                "authors": [
                    {
                        "first": "Richard",
                        "middle": [],
                        "last": "Zens",
                        "suffix": ""
                    },
                    {
                        "first": "Hermann",
                        "middle": [],
                        "last": "Ney",
                        "suffix": ""
                    },
                    {
                        "first": "Taro",
                        "middle": [],
                        "last": "Watanabe",
                        "suffix": ""
                    },
                    {
                        "first": "Eiichiro",
                        "middle": [],
                        "last": "Sumita",
                        "suffix": ""
                    }
                ],
                "year": 2004,
                "venue": "20th International Conference on Computational Linguistics (COLING-2004)",
                "volume": "",
                "issue": "",
                "pages": "",
                "other_ids": {},
                "num": null,
                "urls": [],
                "raw_text": "Richard Zens, Hermann Ney, Taro Watanabe, and Eiichiro Sumita. Reordering constraints for phrase-based statistical machine translation. In 20th International Conference on Computational Linguistics (COLING-2004), Geneva, August 2004.",
                "links": null
            }
        },
        "ref_entries": {
            "TABREF0": {
                "html": null,
                "type_str": "table",
                "num": null,
                "text": "Evaluation results on the IWSLT06 dataset: integrating the WSD translation predictions improves BLEU, NIST, METEOR, WER, PER, CDER and TER across all 3 different available test sets.",
                "content": "<table><tr><td>Test</td><td>Exper.</td><td colspan=\"2\">BLEU NIST METEOR METEOR</td><td>TER</td><td>WER PER</td><td>CDER</td></tr><tr><td>Set</td><td/><td/><td>(no syn)</td><td/><td/></tr><tr><td colspan=\"2\">Test 1 SMT</td><td>42.21 7.888 65.40</td><td>63.24</td><td colspan=\"3\">40.45 45.58 37.80 40.09</td></tr><tr><td/><td colspan=\"2\">SMT+WSD 42.38 7.902 65.73</td><td>63.64</td><td colspan=\"3\">39.98 45.30 37.60 39.91</td></tr><tr><td colspan=\"2\">Test 2 SMT</td><td>41.49 8.167 66.25</td><td>63.85</td><td colspan=\"3\">40.95 46.42 37.52 40.35</td></tr><tr><td/><td colspan=\"2\">SMT+WSD 41.97 8.244 66.35</td><td>63.86</td><td colspan=\"3\">40.63 46.14 37.25 40.10</td></tr><tr><td colspan=\"2\">Test 3 SMT</td><td>49.91 9.016 73.36</td><td>70.70</td><td colspan=\"3\">35.60 40.60 32.30 35.46</td></tr><tr><td/><td colspan=\"2\">SMT+WSD 51.05 9.142 74.13</td><td>71.44</td><td colspan=\"3\">34.68 39.75 31.71 34.58</td></tr></table>"
            },
            "TABREF1": {
                "html": null,
                "type_str": "table",
                "num": null,
                "text": "Evaluation results on the NIST test set: integrating the WSD translation predictions improves BLEU, NIST, METEOR, WER, PER, CDER and TER",
                "content": "<table><tr><td>Exper.</td><td colspan=\"2\">BLEU NIST METEOR METEOR</td><td>TER</td><td>WER PER</td><td>CDER</td></tr><tr><td/><td/><td>(no syn)</td><td/><td/></tr><tr><td>SMT</td><td>20.41 7.155 60.21</td><td>56.15</td><td colspan=\"3\">76.76 88.26 61.71 70.32</td></tr><tr><td colspan=\"2\">SMT+WSD 20.92 7.468 60.30</td><td>56.79</td><td colspan=\"3\">71.34 83.87 57.29 67.38</td></tr><tr><td>5.1 Data set</td><td/><td/><td/><td/></tr></table>"
            },
            "TABREF2": {
                "html": null,
                "type_str": "table",
                "num": null,
                "text": "Translation examples with and without WSD for SMT, drawn from IWSLT data sets.Input \u00f7 l X -. \u00bf Ref.Please transfer to the Chuo train line. SMT Please turn to the Central Line. SMT+WSD Please transfer to Central Line.",
                "content": "<table><tr><td>Input</td><td colspan=\"2\">fh ( f p</td><td/></tr><tr><td>Ref.</td><td colspan=\"3\">Do I pay on the bus?</td></tr><tr><td>SMT</td><td colspan=\"3\">Please get on the bus?</td></tr><tr><td colspan=\"4\">SMT+WSD I buy a ticket on the bus?</td></tr><tr><td>Input</td><td>\u2022 \"\u00a2</td><td/><td/></tr><tr><td>Ref.</td><td colspan=\"3\">Do I need a reservation?</td></tr><tr><td>SMT</td><td colspan=\"3\">I need a reservation?</td></tr><tr><td colspan=\"4\">SMT+WSD Do I need a reservation?</td></tr><tr><td>Input</td><td colspan=\"2\">\u00f3 \u2022 n\u00a4</td><td>\u00d9</td><td>h \" \"\u00a2</td></tr><tr><td>Ref.</td><td colspan=\"4\">I want to reconfirm this ticket.</td></tr><tr><td>SMT</td><td colspan=\"4\">I would like to reconfirm a flight for this ticket.</td></tr><tr><td colspan=\"5\">SMT+WSD I would like to reconfirm my reservation for this ticket.</td></tr><tr><td>Input</td><td colspan=\"2\">eL \u00ef\u00e5 0 \u00a3\u00cc</td><td/></tr><tr><td>Ref.</td><td colspan=\"3\">Can I get there on foot?</td></tr><tr><td>SMT</td><td colspan=\"2\">Is there on foot?</td><td/></tr><tr><td colspan=\"4\">SMT+WSD Can I get there on foot?</td></tr><tr><td>Input</td><td>ae</td><td>* \u00a6</td><td/><td>@\u00e5 \u00f7 \u00eb\u00b9</td></tr><tr><td>Ref.</td><td colspan=\"4\">I have another appointment, so please hurry.</td></tr><tr><td>SMT</td><td colspan=\"4\">I have an appointment for a, so please hurry.</td></tr><tr><td colspan=\"5\">SMT+WSD I have another appointment, so please hurry.</td></tr><tr><td>Input</td><td colspan=\"3\">\u00f9 w `\u00fd J\u00c9</td><td>0~ G \" \u00ef</td></tr><tr><td>Ref.</td><td colspan=\"4\">Excuse me. Could you tell me the way to Broadway?</td></tr><tr><td>SMT</td><td colspan=\"4\">Could you tell me the way to Broadway? I am sorry.</td></tr><tr><td colspan=\"5\">SMT+WSD Excuse me, could you tell me the way to Broadway?</td></tr><tr><td>Input</td><td>\u00f9 w</td><td>\u00f3</td><td colspan=\"2\">* &amp; 7</td></tr><tr><td>Ref.</td><td colspan=\"4\">Excuse me, I want to open an account.</td></tr><tr><td>SMT</td><td colspan=\"4\">Excuse me, I would like to have an account.</td></tr><tr><td colspan=\"5\">SMT+WSD Excuse me, I would like to open an account.</td></tr></table>"
            },
            "TABREF3": {
                "html": null,
                "type_str": "table",
                "num": null,
                "text": "Translation examples with and without WSD for SMT, drawn from the NIST test set. Input \u00a1 \u00fbU \u00aeX \u2022h \u00cd\u00f9 \u00d6 SMT Without any congressmen voted against him. SMT+WSD No congressmen voted against him. Input \u00c4 ( f\u00e3 \u017eL \" ?V \u00e5\u00ca \u00f9 \u00ec T S \u00bb\u00fd \" \u00a6 \u00f4 / \u00e4 \u017d\u00fd \u00c5\u00e7 SMTRussia's policy in Chechnya and CIS neighbors attitude is even more worried that the United States. SMT+WSD Russia's policy in Chechnya and its attitude toward its CIS neighbors cause the United States still more anxiety.",
                "content": "<table><tr><td>Input</td><td>\u00f3\u017d \u017d\u00fd \" \u00baC  \u00b6\u00b5 b</td><td/></tr><tr><td>SMT</td><td>As for the U.S. human rights conditions?</td><td/></tr><tr><td colspan=\"3\">SMT+WSD As for the human rights situation in the U.S.?</td></tr><tr><td>Input</td><td>\u00c2 \u00dc / : \u2020 HB \u00e5, \" OEs</td><td>Ac</td></tr><tr><td>SMT</td><td/><td/></tr></table>"
            }
        }
    }
}