File size: 106,613 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 |
{
"paper_id": "2021",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T01:13:09.952585Z"
},
"title": "Moses and the Character-Based Random Babbling Baseline: CoAStaL at AmericasNLP 2021 Shared Task",
"authors": [
{
"first": "Marcel",
"middle": [],
"last": "Bollmann",
"suffix": "",
"affiliation": {},
"email": "marcel@di.ku.dk"
},
{
"first": "Rahul",
"middle": [],
"last": "Aralikatte",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "H\u00e9ctor",
"middle": [],
"last": "Ricardo",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Murrieta",
"middle": [],
"last": "Bello",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Daniel",
"middle": [],
"last": "Hershcovich",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Anders",
"middle": [],
"last": "S\u00f8gaard",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "University of Copenhagen",
"location": {}
},
"email": "soegaard@di.ku.dk"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "We evaluated a range of neural machine translation techniques developed specifically for low-resource scenarios. Unsuccessfully. In the end, we submitted two runs: (i) a standard phrase-based model, and (ii) a random babbling baseline using character trigrams. We found that it was surprisingly hard to beat (i), in spite of this model being, in theory, a bad fit for polysynthetic languages; and more interestingly, that (ii) was better than several of the submitted systems, highlighting how difficult low-resource machine translation for polysynthetic languages is.",
"pdf_parse": {
"paper_id": "2021",
"_pdf_hash": "",
"abstract": [
{
"text": "We evaluated a range of neural machine translation techniques developed specifically for low-resource scenarios. Unsuccessfully. In the end, we submitted two runs: (i) a standard phrase-based model, and (ii) a random babbling baseline using character trigrams. We found that it was surprisingly hard to beat (i), in spite of this model being, in theory, a bad fit for polysynthetic languages; and more interestingly, that (ii) was better than several of the submitted systems, highlighting how difficult low-resource machine translation for polysynthetic languages is.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "Shared tasks on machine translation are often conducted on large parallel training corpora: for example, the majority of datasets used in the WMT20 shared tasks have sentence pairs in the hundred thousands, often even millions (Barrault et al., 2020) . In contrast, the AmericasNLP 2021 shared task (Mager et al., 2021) provided us with as little as 3,883 sentence pairs (for Ashaninka), and with the exception of Quechua (125k pairs), all languages had fewer than 30k sentence pairs. Additionally, many of these languages are polysynthetic, which is known to provide additional challenges for machine translation (Klavans et al., 2018; Mager et al., 2018b) .",
"cite_spans": [
{
"start": 227,
"end": 250,
"text": "(Barrault et al., 2020)",
"ref_id": null
},
{
"start": 299,
"end": 319,
"text": "(Mager et al., 2021)",
"ref_id": "BIBREF23"
},
{
"start": 614,
"end": 636,
"text": "(Klavans et al., 2018;",
"ref_id": "BIBREF18"
},
{
"start": 637,
"end": 657,
"text": "Mager et al., 2018b)",
"ref_id": "BIBREF22"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "We initially focused our efforts on two areas: (i) obtaining more data, both parallel and monolingual (Sec. 2); and (ii) exploring a range of different neural machine translation techniques, particular those specifically developed for low-resource scenarios, to find a promising system to build on and tweak further. Unfortunately, we were wholly unsuccessful in the latter (Sec. 5). All neural models that we tried performed extremely poorly when compared to a standard statistical phrase-based model (Sec. 3.1). The overall low performance of all our models further prompted us to implement a \"random babbling\" baseline (Sec. 3.2): a model that outputs plausible-looking n-grams in the target language without any actual relation to the source sentences. This baseline, together with the phrasebased model, were the only two systems we ended up submitting. Our main findings are:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "\u2022 It was surprisingly hard to beat a standard phrase-based model, as evidenced not only by our own failed attempts, but also by this system taking third place on three languages in the official evaluation (track 1).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "\u2022 It is apparently challenging for many MT systems to even produce well-formed outputs in the target languages, as our random babbling baseline outperformed at least one other system on nine of the languages, and even took fifth place out of 12 on Ashaninka (track 2).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "We train models for all languages provided by the shared task, using their official training datasets (cf. Table 1 ). As the shared task allowed for using external datasets, we also tried to find more data sources to use for model training.",
"cite_spans": [],
"ref_spans": [
{
"start": 107,
"end": 114,
"text": "Table 1",
"ref_id": "TABREF1"
}
],
"eq_spans": [],
"section": "Data",
"sec_num": "2"
},
{
"text": "We gathered parallel Spanish-totarget datasets for the following languages which should not overlap with the data provided by the shared task organizers: Aymara from JW300 (Agi\u0107 and Vuli\u0107, 2019) ; Guarani from Tatoeba; and Nahuatl and Quechua from the Bible corpus by Christodouloupoulos and Steedman (2015) . We note that for the Bible corpus, the Nahuatl portion is from a narrower dialectal region (NHG \"Tetelcingo Nahuatl\") than the data in the shared task, and it also covers a different variant of Quechua (QUW \"Kichwa\" vs. QUY \"Ayacucho Quechua\"), but we hoped that in this extremely low-resource scenario, this would still prove useful. All datasets were obtained from OPUS 1 (Tiedemann, 2012) .",
"cite_spans": [
{
"start": 172,
"end": 194,
"text": "(Agi\u0107 and Vuli\u0107, 2019)",
"ref_id": "BIBREF0"
},
{
"start": 268,
"end": 307,
"text": "Christodouloupoulos and Steedman (2015)",
"ref_id": "BIBREF9"
},
{
"start": 684,
"end": 701,
"text": "(Tiedemann, 2012)",
"ref_id": "BIBREF35"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Parallel data",
"sec_num": null
},
{
"text": "Monolingual data Wikipedias exist for Aymara, Guaran\u00ed, Nahuatl, and Quechua. We use WikiExtractor (Attardi, 2015) to obtain text data from their respective dumps, 2 then use a small set of regular expressions to clean them from XML tags and entities. This gives us between 28k and 100k lines of text per language. We obtain further monolingual data from several online sources in PDF format. For Nahuatl and H\u00f1\u00e4h\u00f1u, we use a book provided by the Mexican government; 3 for Quechua, we use two books: The Little Prince (Saint-Exup\u00e9ry, 2018) and Antonio Raimondi's Once upon a time.. in Peru (Villacorta, 2007) . The Mexican government also publishes the series Languages from Mexico which contains books based on short stories in Nahuatl (Gustavo et al., 2007) , Raramuri (Arvizu Castillo, 2002) , H\u00f1\u00e4h\u00f1u (Mondrag\u00f3n et al., 2002b) , and Wix\u00e1rika (Mondrag\u00f3n et al., 2002a) . Finally, we also use the Bible translated to Quechua, Guarani, and Aymara. We extract the text for all of these resources with the Google OCR API. 4",
"cite_spans": [
{
"start": 98,
"end": 113,
"text": "(Attardi, 2015)",
"ref_id": null
},
{
"start": 589,
"end": 607,
"text": "(Villacorta, 2007)",
"ref_id": "BIBREF36"
},
{
"start": 736,
"end": 758,
"text": "(Gustavo et al., 2007)",
"ref_id": "BIBREF15"
},
{
"start": 778,
"end": 793,
"text": "Castillo, 2002)",
"ref_id": "BIBREF1"
},
{
"start": 803,
"end": 828,
"text": "(Mondrag\u00f3n et al., 2002b)",
"ref_id": "BIBREF26"
},
{
"start": 844,
"end": 869,
"text": "(Mondrag\u00f3n et al., 2002a)",
"ref_id": "BIBREF25"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Parallel data",
"sec_num": null
},
{
"text": "We first describe the two models we submitted: a standard phrase-based model (CoAStaL-1) and a random babbling baseline (CoAStaL-2). Other models that we experimented with but did not submit for evaluation are discussed later in Sec. 5.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Models",
"sec_num": "3"
},
{
"text": "We train a statistical phrase-based model with Moses (Koehn et al., 2007) using default settings, following the guidelines for training a baseline. 5 We do minimal preprocessing: we use the provided cleaning script and rely on plain whitespace tokenization, with the only exception that we also insert spaces around square brackets. The language model is trained with 5-grams instead of 3-grams, as this improved the results very slightly on the development sets. We train a separate model for each language and use the respective development set for tuning before translating the test set.",
"cite_spans": [
{
"start": 53,
"end": 73,
"text": "(Koehn et al., 2007)",
"ref_id": "BIBREF19"
},
{
"start": 148,
"end": 149,
"text": "5",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Phrase-Based MT",
"sec_num": "3.1"
},
{
"text": "The models we submitted did, mistakenly, not make use of the additional parallel data we gathered (cf. Sec. 2). We evaluated the same system trained with this additional data after the deadline, but unfortunately did not observe an improvement; we present results for both variants in Sec. 4.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Phrase-Based MT",
"sec_num": "3.1"
},
{
"text": "Since we observed very low scores for all the models we tried, we wanted to compare with a baseline that generates text based only on (i) n-gram distributions in the target language, and (ii) lengths of the source sentences. We call this baseline random babbling because it is in no way conditioned on the actual words in the source sentences.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Random Babbling Baseline",
"sec_num": "3.2"
},
{
"text": "Concretely, we \"train\" our baseline by extracting and counting all character trigrams in the training file of the target language. Characters were chosen over words as the official evaluation metric of the shared task, chrF, is character-based. We also calculate the average length ratio of the sentence pairs in order to determine the desired length of our \"translation\" at test time. To generate output, we simply choose the top n most frequent character trigrams, with n chosen so that the desired sentence length is reached. 6 Lastly, we perform a few tweaks to disguise this babbling as an actual translation: (i) we randomize the order of the chosen trigrams, (ii) reduce multiple consecutive whitespace characters to a single space, (iii) lowercase all characters that are not word-initial and uppercase the sentence-initial Table 2 : Results for our submitted models on DEV and TEST sets. All TEST results are from the official evaluation except for the \"Phrase-based + extra data\" setting, which we evaluated after the deadline.",
"cite_spans": [
{
"start": 529,
"end": 530,
"text": "6",
"ref_id": null
}
],
"ref_spans": [
{
"start": 832,
"end": 839,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Random Babbling Baseline",
"sec_num": "3.2"
},
{
"text": "character, and (iv) if the sequence does not end in a punctuation mark but the Spanish source sentence did, we copy and add this punctuation character from the source side.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Random Babbling Baseline",
"sec_num": "3.2"
},
{
"text": "Results of our models are shown in Table 2 , both for our own evaluation on the development sets and for the official evaluation on the test sets (Ebrahimi et al., 2021) .",
"cite_spans": [
{
"start": 146,
"end": 169,
"text": "(Ebrahimi et al., 2021)",
"ref_id": "BIBREF23"
}
],
"ref_spans": [
{
"start": 35,
"end": 42,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Results",
"sec_num": "4"
},
{
"text": "Phrase-Based MT Our phrase-based model (Sec. 3.1) was ranked in track 1 of the shared task evaluation as it makes use of the development sets for tuning. Compared to the other systems evaluated in this track, we observe a solid average performance of our model-it usually ranks in the middle of the field, with the best placement being 3rd on Bribri, H\u00f1\u00e4h\u00f1u, and Shipibo-Konibo, and the worst ranking being 8th out of 11 on Guarani. In terms of chrF score, the model ranges between 0.159 (on Raramuri) and 0.297 (on Shipibo-Konibo), but we note that there is a noticeable gap to the bestperforming system, Helsinki-2, which outperforms ours by about +0.09 chrF on average.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Results",
"sec_num": "4"
},
{
"text": "Random Babbling Our random babbling baseline (Sec. 3.2) did not make use of the development sets and was therefore ranked in track 2 of the official evaluation. Amazingly, it almost never ranks last and even takes 5th place out of 12 on Ashaninka. It also outperforms the official baseline on eight of the languages. In terms of BLEU score, on the other hand, this model usually scores close to zero. This is because we based it on character trigrams; if we wanted to optimize for BLEU, we could have chosen word-based babbling instead.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Results",
"sec_num": "4"
},
{
"text": "Comparing across the tracks with our first, phrasebased system, we observe that the latter scores consistently better, which is reassuring.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Results",
"sec_num": "4"
},
{
"text": "We intended our phrase-based Moses system more as a baseline for our experiments with different neural models than as an actual system submission. It was surprising to us how clearly this system outperformed our attempts at building a neural MT system, and that it already did so with its default configuration. In theory, whitespace tokenization should be a bad fit for polysynthetic languages, as a high degree of morphological complexity exacerbates the problem of data sparsity and rarely seen word forms. We experimented with different subword tokenization techniques in combination with Moses, but this always resulted in degraded performance on the development sets. The random babbling baseline was motivated by two observations: (i) performance was extremely low for all models we tried, and (ii) outputs of the neural models frequently looked very unnatural, to the point that the models had not yet learned how to form plausible-looking sentences in the target languages. This is quite typical behavior for underfitted neural models. As an example, this is an output we observed when running the official baseline system on the development set for Raramuri:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "(1) IN:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "Realmente no me importa si tengo un lugar para vivir. GOLD: Ke chibi ir\u00e9 mapure ke nir\u00falisaka k\u00fami ne bet\u00e9lima. PRED: ( 2 ) ( a ) k\u00e9 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 ne ga'r\u00e1 [. . . ] This prompted us to implement a baseline which, while having no relation to the actual input sentence, at least better resembles the typical distribution of character n-grams in the given language.",
"cite_spans": [
{
"start": 337,
"end": 345,
"text": "[. . . ]",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "Here is an example from the test set for Ashaninka with outputs from both our phrase-based (SYS-1) and random (SYS-2) model:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "(2) IN: Todav\u00eda estoy trabajando para este d\u00eda. GOLD: Irosatitatsi nantabeeti oka kitaiteriki. SYS-1: Tekirata nosaikaki trabajando inchamenta itovantarori.\" d\u00eda. SYS-2: Iritsiri irotakntakanarishiantakiro aka.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "We can see that both system outputs bear very little resemblance to the gold translation or to each other. While Moses (SYS-1) copies a few Spanish words and includes implausibly placed punctuation marks, random babbling (SYS-2) produces output of similar length to the correct translation and overlaps with it in several observable character trigrams (e.g. iro, tsi, ant).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "Obviously, the random babbling baseline is not meant as an actual suggestion for a translation system-it literally does not \"translate\" anything. However, as the official shared task evaluation and the examples above show, it can serve as a useful \"sanity check\" for situations where the performance of actual MT systems is so low that it is unclear whether they even acquired superficial knowledge of character distributions in the target language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "4.1"
},
{
"text": "Here we briefly describe other ideas that we pursued, but were unfortunately not successful with, so we did not submit any systems based on these techniques for evaluation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Things that did not work",
"sec_num": "5"
},
{
"text": "Pre-trained Transformers Following Rothe et al. 2020, we use an auto-encoding transformer as the encoder and an auto-regressive transformer as the decoder of a sequence-to-sequence model. Out of the several configurations we experimented with, the best performance was observed when the encoder is pre-trained on the Spanish OSCAR corpus (Ortiz Su\u00e1rez et al., 2020) and the decoders are pre-trained on language-specific monolingual corpora collected from the web (cf. Sec. 2) along with the target files of the training data. However, the results were not on-par with the simpler models; averaging over all languages, we observed a chrF score of 0.12 on the dev sets, compared to 0.23 with the phrase-based model (cf. Sec. 3.1). We postulate that the training data was just not enough to train the cross-attention weights between the encoder and decoders. Note that these weights need to be trained from scratch, as opposed to the other weights which are initialized from language modelling checkpoints.",
"cite_spans": [
{
"start": 338,
"end": 365,
"text": "(Ortiz Su\u00e1rez et al., 2020)",
"ref_id": "BIBREF29"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Things that did not work",
"sec_num": "5"
},
{
"text": "In an attempt to improve the transformer-based models, we used the shared task data to train similar transformer-based models in the reverse direction, i.e. to Spanish, in order to back-translate the monolingual corpora (cf. Sec. 2). This would give us automatically translated Spanish outputs to use as the source side for additional training data (Sennrich et al., 2016; Hoang et al., 2018) . Since monolingual data in Spanish-which was used to pre-train the decoder's language model for this experiment-is abundant, we expected the machine-translated Spanish text to be of reasonably good quality. However, the models turned out to perform quite badly, with the resulting Spanish text being of very low quality and often very repetitive. We therefore decided to abandon this direction after preliminary experiments.",
"cite_spans": [
{
"start": 349,
"end": 372,
"text": "(Sennrich et al., 2016;",
"ref_id": "BIBREF34"
},
{
"start": 373,
"end": 392,
"text": "Hoang et al., 2018)",
"ref_id": "BIBREF17"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Back-translation",
"sec_num": null
},
{
"text": "Character-Level NMT Since many of the languages in the shared task are polysynthetic, a character-level model might be better suited here, as it can better learn morphology (Belinkov et al., 2017) . We train fully character-level models following Lee et al. (2017) , which are based on com-bining convolutional and recurrent layers. 7 Finding a good hyperparameter configuration for this model proved very time-consuming; the best configuration we found modifies the original model by using half the number of units in the embedding layer and decoder layers (256 and 512, respectively). For Quechua, which we initially experimented on, this yielded a chrF score of 0.33 on the dev set vs. 0.27 with phrase-based MT, but we ran out of time to train models for the other languages. A post-hoc evaluation on the other languages failed to replicate this success, though. Potentially, the hyperparameter configuration is very sensitive to the language in question, or the amount of training data was not enough for the other languages (Quechua had by far the largest training set of all languages in the shared task).",
"cite_spans": [
{
"start": 173,
"end": 196,
"text": "(Belinkov et al., 2017)",
"ref_id": "BIBREF6"
},
{
"start": 247,
"end": 264,
"text": "Lee et al. (2017)",
"ref_id": "BIBREF20"
},
{
"start": 333,
"end": 334,
"text": "7",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Back-translation",
"sec_num": null
},
{
"text": "We train NMT models using a language model prior, following Baziotis et al. (2020) . This method allows us to make use of the additional monolingual data we gathered (cf. Sec. 2) within a neural MT framework, and we hoped that this would help the model to produce valid words in the target languages, i.e., reduce the \"babbling\" effect we saw in outputs like Example (1) above. We focused our efforts on the LSTM-based models provided by the authors 8 rather than the transformer ones, since we believe that those should be easier to train in this extremely low-resource setting. Despite experimenting with different hyperparameters (including number and size of LSTM layers), we could not exceed an average 0.16 chrF on the dev sets (compared to 0.23 with the phrase-based model).",
"cite_spans": [
{
"start": 60,
"end": 82,
"text": "Baziotis et al. (2020)",
"ref_id": "BIBREF5"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Language Model Prior",
"sec_num": null
},
{
"text": "We experiment with graph convolutional encoders using the framework by Bastings et al. (2017) . Thus, we train NMT systems that operate directly over graphs; in our case, syntactic annotations of the source sentences following the Universal Dependencies (UD) scheme (Nivre et al., 2020) . We parsed the all the source sentences from training set provided by the task organizer with Stanza (Qi et al., 2020) . We were initially motivated to follow this approach because UD annotation can provide extra information to the encoder to generate better translations, ideally with less data. Even though we tested several configurations, not even our best architecture-two layers of GCN encoder with 250 units, and LSTM decoder with 250 units, trained for 5 epochs, with a vocabulary of 5000 words in source and targetwas able to outperform the random babbling system. We hypothesize that with this amount of examples, UD's external information is not sufficient to produce an efficient encoder.",
"cite_spans": [
{
"start": 71,
"end": 93,
"text": "Bastings et al. (2017)",
"ref_id": "BIBREF4"
},
{
"start": 266,
"end": 286,
"text": "(Nivre et al., 2020)",
"ref_id": "BIBREF27"
},
{
"start": 389,
"end": 406,
"text": "(Qi et al., 2020)",
"ref_id": "BIBREF31"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Graph Convolutional Encoders",
"sec_num": null
},
{
"text": "The (relative) success of our random babbling baseline shows that many MT systems fail to reproduce even superficial characteristics of word formation and character distribution in the target languages; a result that was confirmed by our own failed attempts at training a competitive neural MT model.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "Out of the neural models we tried, purely character-level MT was among the more promising ones. We speculate that in the Spanish-to-target setting, a model that combines a strong pre-trained Spanish encoder with a purely character-level decoder might be a promising direction for further experiments.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "We also note that there are several languagespecific resources, such as morphological segmentation tools, 9 that might be worth using. We focused our efforts here on finding a broadly applicable architecture without any language-specific components, but would be curious to see if including such components can yield significant improvements on individual languages.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "https://opus.nlpl.eu/ 2 https://dumps.wikimedia.org/ 3 https://www.gob.mx/inpi/documentos/ libros-en-lenguas-indigenas 4 https://cloud.google.com/vision/docs/ pdf",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "http://www.statmt.org/moses/?n=Moses. Baseline6 We also tried random baseline models with other n-gram lengths, sampling from the distribution (instead of always picking the most frequent items), and training a simple language model, but found nothing that significantly improved on this approach on the development set.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "We use our own reimplementation of the authors' code. 8 https://github.com/cbaziotis/ lm-prior-for-nmt",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "Marcel Bollmann was funded from the European Union's Horizon 2020 research and innovation programme under the Marie Sk\u0142odowska-Curie grant agreement No. 845995. Rahul Aralikatte and Anders S\u00f8gaard were funded by a Google Focused Research Award. Miryam de Lhoneux was funded by the Swedish Research Council (grant 2020-00437).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgements",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "JW300: A widecoverage parallel corpus for low-resource languages",
"authors": [
{
"first": "\u017deljko",
"middle": [],
"last": "Agi\u0107",
"suffix": ""
},
{
"first": "Ivan",
"middle": [],
"last": "Vuli\u0107",
"suffix": ""
}
],
"year": 2019,
"venue": "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
"volume": "",
"issue": "",
"pages": "3204--3210",
"other_ids": {
"DOI": [
"10.18653/v1/P19-1310"
]
},
"num": null,
"urls": [],
"raw_text": "\u017deljko Agi\u0107 and Ivan Vuli\u0107. 2019. JW300: A wide- coverage parallel corpus for low-resource languages. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3204-3210, Florence, Italy. Association for Compu- tational Linguistics.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Relatos tarahumaras = Ki'\u00e1 ra'ichaala rar\u00e1muli. CNCA-Direcci\u00f3n Gen-9 e.g. Apertium for Guarani",
"authors": [
{
"first": "Teresa Arvizu",
"middle": [],
"last": "Castillo",
"suffix": ""
}
],
"year": 2002,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Teresa Arvizu Castillo. 2002. Relatos tarahumaras = Ki'\u00e1 ra'ichaala rar\u00e1muli. CNCA-Direcci\u00f3n Gen- 9 e.g. Apertium for Guarani: https://github.com/ apertium/apertium-grn eral de Culturas Populares e Ind\u00edgenas, Ciudad de Mexico.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Proceedings of the Fifth Conference on Machine Translation",
"authors": [
{
"first": "Lo\u00efc",
"middle": [],
"last": "Barrault",
"suffix": ""
},
{
"first": "Magdalena",
"middle": [],
"last": "Biesialska",
"suffix": ""
},
{
"first": "Ond\u0159ej",
"middle": [],
"last": "Bojar",
"suffix": ""
},
{
"first": "Marta",
"middle": [
"R"
],
"last": "Costa-Juss\u00e0",
"suffix": ""
},
{
"first": "Christian",
"middle": [],
"last": "Federmann",
"suffix": ""
},
{
"first": "Yvette",
"middle": [],
"last": "Graham",
"suffix": ""
},
{
"first": "Roman",
"middle": [],
"last": "Grundkiewicz",
"suffix": ""
},
{
"first": "Barry",
"middle": [],
"last": "Haddow",
"suffix": ""
},
{
"first": "Matthias",
"middle": [],
"last": "Huck",
"suffix": ""
},
{
"first": "Eric",
"middle": [],
"last": "Joanis",
"suffix": ""
},
{
"first": "Tom",
"middle": [],
"last": "Kocmi",
"suffix": ""
},
{
"first": "Philipp",
"middle": [],
"last": "Koehn",
"suffix": ""
},
{
"first": "Chi-Kiu",
"middle": [],
"last": "Lo",
"suffix": ""
},
{
"first": "Nikola",
"middle": [],
"last": "Ljube\u0161i\u0107",
"suffix": ""
},
{
"first": "Christof",
"middle": [],
"last": "Monz",
"suffix": ""
},
{
"first": "Makoto",
"middle": [],
"last": "Morishita",
"suffix": ""
},
{
"first": "Masaaki",
"middle": [],
"last": "Nagata",
"suffix": ""
},
{
"first": "Toshiaki",
"middle": [],
"last": "Nakazawa",
"suffix": ""
}
],
"year": null,
"venue": "",
"volume": "",
"issue": "",
"pages": "1--55",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lo\u00efc Barrault, Magdalena Biesialska, Ond\u0159ej Bojar, Marta R. Costa-juss\u00e0, Christian Federmann, Yvette Graham, Roman Grundkiewicz, Barry Haddow, Matthias Huck, Eric Joanis, Tom Kocmi, Philipp Koehn, Chi-kiu Lo, Nikola Ljube\u0161i\u0107, Christof Monz, Makoto Morishita, Masaaki Nagata, Toshi- aki Nakazawa, Santanu Pal, Matt Post, and Marcos Zampieri. 2020. Findings of the 2020 conference on machine translation (WMT20). In Proceedings of the Fifth Conference on Machine Translation, pages 1-55, Online. Association for Computational Lin- guistics.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Graph convolutional encoders for syntax-aware neural machine translation",
"authors": [
{
"first": "Jasmijn",
"middle": [],
"last": "Bastings",
"suffix": ""
},
{
"first": "Ivan",
"middle": [],
"last": "Titov",
"suffix": ""
},
{
"first": "Wilker",
"middle": [],
"last": "Aziz",
"suffix": ""
},
{
"first": "Diego",
"middle": [],
"last": "Marcheggiani",
"suffix": ""
},
{
"first": "Khalil",
"middle": [],
"last": "Sima",
"suffix": ""
}
],
"year": 2017,
"venue": "Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing",
"volume": "",
"issue": "",
"pages": "1957--1967",
"other_ids": {
"DOI": [
"10.18653/v1/D17-1209"
]
},
"num": null,
"urls": [],
"raw_text": "Jasmijn Bastings, Ivan Titov, Wilker Aziz, Diego Marcheggiani, and Khalil Sima'an. 2017. Graph convolutional encoders for syntax-aware neural ma- chine translation. In Proceedings of the 2017 Con- ference on Empirical Methods in Natural Language Processing, pages 1957-1967, Copenhagen, Den- mark. Association for Computational Linguistics.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Language model prior for low-resource neural machine translation",
"authors": [
{
"first": "Christos",
"middle": [],
"last": "Baziotis",
"suffix": ""
},
{
"first": "Barry",
"middle": [],
"last": "Haddow",
"suffix": ""
},
{
"first": "Alexandra",
"middle": [],
"last": "Birch",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
"volume": "",
"issue": "",
"pages": "7622--7634",
"other_ids": {
"DOI": [
"10.18653/v1/2020.emnlp-main.615"
]
},
"num": null,
"urls": [],
"raw_text": "Christos Baziotis, Barry Haddow, and Alexandra Birch. 2020. Language model prior for low-resource neu- ral machine translation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Lan- guage Processing (EMNLP), pages 7622-7634, On- line. Association for Computational Linguistics.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "What do neural machine translation models learn about morphology?",
"authors": [
{
"first": "Yonatan",
"middle": [],
"last": "Belinkov",
"suffix": ""
},
{
"first": "Nadir",
"middle": [],
"last": "Durrani",
"suffix": ""
},
{
"first": "Fahim",
"middle": [],
"last": "Dalvi",
"suffix": ""
},
{
"first": "Hassan",
"middle": [],
"last": "Sajjad",
"suffix": ""
},
{
"first": "James",
"middle": [],
"last": "Glass",
"suffix": ""
}
],
"year": 2017,
"venue": "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics",
"volume": "",
"issue": "",
"pages": "861--872",
"other_ids": {
"DOI": [
"10.18653/v1/P17-1080"
]
},
"num": null,
"urls": [],
"raw_text": "Yonatan Belinkov, Nadir Durrani, Fahim Dalvi, Has- san Sajjad, and James Glass. 2017. What do neu- ral machine translation models learn about morphol- ogy? In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Vol- ume 1: Long Papers), pages 861-872, Vancouver, Canada. Association for Computational Linguistics.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Diccionario Raramuri-Castellano (Tarahumara)",
"authors": [
{
"first": "David",
"middle": [],
"last": "Brambila",
"suffix": ""
}
],
"year": 1976,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "David Brambila. 1976. Diccionario Raramuri- Castellano (Tarahumara). Obra Nacional de la Buena Prensa, Mexico.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Development of a Guarani -Spanish parallel corpus",
"authors": [
{
"first": "Luis",
"middle": [],
"last": "Chiruzzo",
"suffix": ""
},
{
"first": "Pedro",
"middle": [],
"last": "Amarilla",
"suffix": ""
},
{
"first": "Adolfo",
"middle": [],
"last": "R\u00edos",
"suffix": ""
},
{
"first": "Gustavo",
"middle": [
"Gim\u00e9nez"
],
"last": "Lugo",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 12th Language Resources and Evaluation Conference",
"volume": "",
"issue": "",
"pages": "2629--2633",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Luis Chiruzzo, Pedro Amarilla, Adolfo R\u00edos, and Gus- tavo Gim\u00e9nez Lugo. 2020. Development of a Guarani -Spanish parallel corpus. In Proceedings of the 12th Language Resources and Evaluation Con- ference, pages 2629-2633, Marseille, France. Euro- pean Language Resources Association.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "A massively parallel corpus: the Bible in 100 languages. Language Resources and Evaluation",
"authors": [
{
"first": "Christos",
"middle": [],
"last": "Christodouloupoulos",
"suffix": ""
},
{
"first": "Mark",
"middle": [],
"last": "Steedman",
"suffix": ""
}
],
"year": 2015,
"venue": "",
"volume": "49",
"issue": "",
"pages": "375--395",
"other_ids": {
"DOI": [
"10.1007/s10579-014-9287-y"
]
},
"num": null,
"urls": [],
"raw_text": "Christos Christodouloupoulos and Mark Steedman. 2015. A massively parallel corpus: the Bible in 100 languages. Language Resources and Evalua- tion, 49(2):375-395.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "Tsunkua -corpus paralelo otom\u00ed-espa\u00f1ol",
"authors": [
{
"first": "Comunidad",
"middle": [],
"last": "Elotl",
"suffix": ""
}
],
"year": 2021,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Comunidad Elotl. 2021. Tsunkua -corpus paralelo otom\u00ed-espa\u00f1ol.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "\u00d1aantsipeta ash\u00e1ninkaki birakochaki. Diccionario Ash\u00e1ninka-Castellano. Versi\u00f3n preliminar",
"authors": [
{
"first": "Cushimariano",
"middle": [],
"last": "Rub\u00e9n",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Romano",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Richer",
"suffix": ""
},
{
"first": "Q",
"middle": [],
"last": "Sebasti\u00e1n",
"suffix": ""
}
],
"year": 2008,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Rub\u00e9n Cushimariano Romano and Richer C. Se- basti\u00e1n Q. 2008. \u00d1aantsipeta ash\u00e1ninkaki bi- rakochaki. Diccionario Ash\u00e1ninka-Castellano. Ver- si\u00f3n preliminar. http://www.lengamer. org/publicaciones/diccionarios/.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Americasnli: Evaluating zero-shot natural language understanding of pretrained multilingual models",
"authors": [
{
"first": "Abteen",
"middle": [],
"last": "Ebrahimi",
"suffix": ""
},
{
"first": "Manuel",
"middle": [],
"last": "Mager",
"suffix": ""
},
{
"first": "Arturo",
"middle": [],
"last": "Oncevay",
"suffix": ""
},
{
"first": "Vishrav",
"middle": [],
"last": "Chaudhary",
"suffix": ""
},
{
"first": "Luis",
"middle": [],
"last": "Chiruzzo",
"suffix": ""
},
{
"first": "Angela",
"middle": [],
"last": "Fan",
"suffix": ""
},
{
"first": "John",
"middle": [],
"last": "Ortega",
"suffix": ""
},
{
"first": "Ricardo",
"middle": [],
"last": "Ramos",
"suffix": ""
},
{
"first": "Annette",
"middle": [],
"last": "Rios",
"suffix": ""
},
{
"first": "Ivan",
"middle": [],
"last": "Vladimir",
"suffix": ""
},
{
"first": "Gustavo",
"middle": [
"A"
],
"last": "Gim\u00e9nez-Lugo",
"suffix": ""
},
{
"first": "Elisabeth",
"middle": [],
"last": "Mager",
"suffix": ""
}
],
"year": null,
"venue": "Ngoc Thang Vu, and Katharina Kann. 2021",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Abteen Ebrahimi, Manuel Mager, Arturo Oncevay, Vishrav Chaudhary, Luis Chiruzzo, Angela Fan, John Ortega, Ricardo Ramos, Annette Rios, Ivan Vladimir, Gustavo A. Gim\u00e9nez-Lugo, Elisabeth Mager, Graham Neubig, Alexis Palmer, Rolando A. Coto Solano, Ngoc Thang Vu, and Katharina Kann. 2021. Americasnli: Evaluating zero-shot nat- ural language understanding of pretrained multilin- gual models in truly low-resource languages.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Neural machine translation models with back-translation for the extremely low-resource indigenous language Bribri",
"authors": [
{
"first": "Isaac",
"middle": [],
"last": "Feldman",
"suffix": ""
},
{
"first": "Rolando",
"middle": [],
"last": "Coto-Solano",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 28th International Conference on Computational Linguistics",
"volume": "",
"issue": "",
"pages": "3965--3976",
"other_ids": {
"DOI": [
"10.18653/v1/2020.coling-main.351"
]
},
"num": null,
"urls": [],
"raw_text": "Isaac Feldman and Rolando Coto-Solano. 2020. Neu- ral machine translation models with back-translation for the extremely low-resource indigenous language Bribri. In Proceedings of the 28th International Conference on Computational Linguistics, pages 3965-3976, Barcelona, Spain (Online). Interna- tional Committee on Computational Linguistics.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Corpus creation and initial SMT experiments between Spanish and Shipibo-konibo",
"authors": [
{
"first": "Ana-Paula",
"middle": [],
"last": "Galarreta",
"suffix": ""
},
{
"first": "Andr\u00e9s",
"middle": [],
"last": "Melgar",
"suffix": ""
},
{
"first": "Arturo",
"middle": [],
"last": "Oncevay",
"suffix": ""
}
],
"year": 2017,
"venue": "Proceedings of the International Conference Recent Advances in Natural Language Processing",
"volume": "",
"issue": "",
"pages": "238--244",
"other_ids": {
"DOI": [
"10.26615/978-954-452-049-6_033"
]
},
"num": null,
"urls": [],
"raw_text": "Ana-Paula Galarreta, Andr\u00e9s Melgar, and Arturo On- cevay. 2017. Corpus creation and initial SMT ex- periments between Spanish and Shipibo-konibo. In Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017, pages 238-244, Varna, Bulgaria. INCOMA Ltd.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Tepole Rivera Miguel \u00c1ngel, and Tzanahua Antonio",
"authors": [
{
"first": "Gustavo",
"middle": [],
"last": "Aguilar Guti\u00e9rrez",
"suffix": ""
},
{
"first": "Arellano",
"middle": [],
"last": "Zamora Rogelio",
"suffix": ""
},
{
"first": "Conde",
"middle": [],
"last": "Reyes Magdaleno",
"suffix": ""
}
],
"year": 2007,
"venue": "Relatos nahuas = Nahua zazanilli. CNCA-Direcci\u00f3n General de Culturas Populares e Ind\u00edgenas",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Aguilar Guti\u00e9rrez Gustavo, Arellano Zamora Rogelio, Conde Reyes Magdaleno, Tepole Rivera Miguel \u00c1n- gel, and Tzanahua Antonio. 2007. Relatos nahuas = Nahua zazanilli. CNCA-Direcci\u00f3n General de Cul- turas Populares e Ind\u00edgenas, Ciudad de Mexico.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Axolotl: a web accessible parallel corpus for Spanish-Nahuatl",
"authors": [
{
"first": "Ximena",
"middle": [],
"last": "Gutierrez-Vasques",
"suffix": ""
},
{
"first": "Gerardo",
"middle": [],
"last": "Sierra",
"suffix": ""
},
{
"first": "Isaac",
"middle": [],
"last": "Hernandez Pompa",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16)",
"volume": "",
"issue": "",
"pages": "4210--4214",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ximena Gutierrez-Vasques, Gerardo Sierra, and Isaac Hernandez Pompa. 2016. Axolotl: a web accessible parallel corpus for Spanish-Nahuatl. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16), pages 4210-4214, Portoro\u017e, Slovenia. European Language Resources Association (ELRA).",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "Iterative backtranslation for neural machine translation",
"authors": [
{
"first": "Duy",
"middle": [],
"last": "Vu Cong",
"suffix": ""
},
{
"first": "Philipp",
"middle": [],
"last": "Hoang",
"suffix": ""
},
{
"first": "Gholamreza",
"middle": [],
"last": "Koehn",
"suffix": ""
},
{
"first": "Trevor",
"middle": [],
"last": "Haffari",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Cohn",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the 2nd Workshop on Neural Machine Translation and Generation",
"volume": "",
"issue": "",
"pages": "18--24",
"other_ids": {
"DOI": [
"10.18653/v1/W18-2703"
]
},
"num": null,
"urls": [],
"raw_text": "Vu Cong Duy Hoang, Philipp Koehn, Gholamreza Haffari, and Trevor Cohn. 2018. Iterative back- translation for neural machine translation. In Pro- ceedings of the 2nd Workshop on Neural Machine Translation and Generation, pages 18-24, Mel- bourne, Australia. Association for Computational Linguistics.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "Challenges in speech recognition and translation of high-value lowdensity polysynthetic languages",
"authors": [
{
"first": "Judith",
"middle": [],
"last": "Klavans",
"suffix": ""
},
{
"first": "John",
"middle": [],
"last": "Morgan",
"suffix": ""
},
{
"first": "Stephen",
"middle": [],
"last": "Larocca",
"suffix": ""
},
{
"first": "Jeffrey",
"middle": [],
"last": "Micher",
"suffix": ""
},
{
"first": "Clare",
"middle": [],
"last": "Voss",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the 13th Conference of the Association for Machine Translation in the Americas",
"volume": "2",
"issue": "",
"pages": "283--293",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Judith Klavans, John Morgan, Stephen LaRocca, Jef- frey Micher, and Clare Voss. 2018. Challenges in speech recognition and translation of high-value low- density polysynthetic languages. In Proceedings of the 13th Conference of the Association for Machine Translation in the Americas (Volume 2: User Track), pages 283-293, Boston, MA. Association for Ma- chine Translation in the Americas.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "Moses: Open source toolkit for statistical machine translation",
"authors": [
{
"first": "Philipp",
"middle": [],
"last": "Koehn",
"suffix": ""
},
{
"first": "Hieu",
"middle": [],
"last": "Hoang",
"suffix": ""
},
{
"first": "Alexandra",
"middle": [],
"last": "Birch",
"suffix": ""
},
{
"first": "Chris",
"middle": [],
"last": "Callison-Burch",
"suffix": ""
},
{
"first": "Marcello",
"middle": [],
"last": "Federico",
"suffix": ""
},
{
"first": "Nicola",
"middle": [],
"last": "Bertoldi",
"suffix": ""
},
{
"first": "Brooke",
"middle": [],
"last": "Cowan",
"suffix": ""
},
{
"first": "Wade",
"middle": [],
"last": "Shen",
"suffix": ""
},
{
"first": "Christine",
"middle": [],
"last": "Moran",
"suffix": ""
},
{
"first": "Richard",
"middle": [],
"last": "Zens",
"suffix": ""
},
{
"first": "Chris",
"middle": [],
"last": "Dyer",
"suffix": ""
},
{
"first": "Ond\u0159ej",
"middle": [],
"last": "Bojar",
"suffix": ""
},
{
"first": "Alexandra",
"middle": [],
"last": "Constantin",
"suffix": ""
},
{
"first": "Evan",
"middle": [],
"last": "Herbst",
"suffix": ""
}
],
"year": 2007,
"venue": "Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics Companion Volume Proceedings of the Demo and Poster Sessions",
"volume": "",
"issue": "",
"pages": "177--180",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Philipp Koehn, Hieu Hoang, Alexandra Birch, Chris Callison-Burch, Marcello Federico, Nicola Bertoldi, Brooke Cowan, Wade Shen, Christine Moran, Richard Zens, Chris Dyer, Ond\u0159ej Bojar, Alexandra Constantin, and Evan Herbst. 2007. Moses: Open source toolkit for statistical machine translation. In Proceedings of the 45th Annual Meeting of the As- sociation for Computational Linguistics Companion Volume Proceedings of the Demo and Poster Ses- sions, pages 177-180, Prague, Czech Republic. As- sociation for Computational Linguistics.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Fully character-level neural machine translation without explicit segmentation",
"authors": [
{
"first": "Jason",
"middle": [],
"last": "Lee",
"suffix": ""
},
{
"first": "Kyunghyun",
"middle": [],
"last": "Cho",
"suffix": ""
},
{
"first": "Thomas",
"middle": [],
"last": "Hofmann",
"suffix": ""
}
],
"year": 2017,
"venue": "Transactions of the Association for Computational Linguistics",
"volume": "5",
"issue": "",
"pages": "365--378",
"other_ids": {
"DOI": [
"10.1162/tacl_a_00067"
]
},
"num": null,
"urls": [],
"raw_text": "Jason Lee, Kyunghyun Cho, and Thomas Hofmann. 2017. Fully character-level neural machine trans- lation without explicit segmentation. Transactions of the Association for Computational Linguistics, 5:365-378.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "Probabilistic finite-state morphological segmenter for wixarika (huichol) language",
"authors": [
{
"first": "Manuel",
"middle": [],
"last": "Mager",
"suffix": ""
},
{
"first": "Di\u00f3nico",
"middle": [],
"last": "Carrillo",
"suffix": ""
},
{
"first": "Ivan",
"middle": [],
"last": "Meza",
"suffix": ""
}
],
"year": 2018,
"venue": "Journal of Intelligent & Fuzzy Systems",
"volume": "34",
"issue": "5",
"pages": "3081--3087",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Manuel Mager, Di\u00f3nico Carrillo, and Ivan Meza. 2018a. Probabilistic finite-state morphological seg- menter for wixarika (huichol) language. Journal of Intelligent & Fuzzy Systems, 34(5):3081-3087.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Lost in translation: Analysis of information loss during machine translation between polysynthetic and fusional languages",
"authors": [
{
"first": "Manuel",
"middle": [],
"last": "Mager",
"suffix": ""
},
{
"first": "Elisabeth",
"middle": [],
"last": "Mager",
"suffix": ""
},
{
"first": "Alfonso",
"middle": [],
"last": "Medina-Urrea",
"suffix": ""
},
{
"first": "Ivan Vladimir Meza",
"middle": [],
"last": "Ruiz",
"suffix": ""
},
{
"first": "Katharina",
"middle": [],
"last": "Kann",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the Workshop on Computational Modeling of Polysynthetic Languages",
"volume": "",
"issue": "",
"pages": "73--83",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Manuel Mager, Elisabeth Mager, Alfonso Medina- Urrea, Ivan Vladimir Meza Ruiz, and Katharina Kann. 2018b. Lost in translation: Analysis of in- formation loss during machine translation between polysynthetic and fusional languages. In Proceed- ings of the Workshop on Computational Modeling of Polysynthetic Languages, pages 73-83, Santa Fe, New Mexico, USA. Association for Computational Linguistics.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "Findings of the AmericasNLP 2021 Shared Task on Open Machine Translation for Indigenous Languages of the Americas",
"authors": [
{
"first": "Manuel",
"middle": [],
"last": "Mager",
"suffix": ""
},
{
"first": "Arturo",
"middle": [],
"last": "Oncevay",
"suffix": ""
},
{
"first": "Abteen",
"middle": [],
"last": "Ebrahimi",
"suffix": ""
},
{
"first": "John",
"middle": [],
"last": "Ortega",
"suffix": ""
},
{
"first": "Annette",
"middle": [],
"last": "Rios",
"suffix": ""
},
{
"first": "Angela",
"middle": [],
"last": "Fan",
"suffix": ""
},
{
"first": "Ximena",
"middle": [],
"last": "Gutierrez-Vasques",
"suffix": ""
},
{
"first": "Luis",
"middle": [],
"last": "Chiruzzo",
"suffix": ""
},
{
"first": "Gustavo",
"middle": [],
"last": "Gim\u00e9nez-Lugo",
"suffix": ""
},
{
"first": "Ricardo",
"middle": [],
"last": "Ramos",
"suffix": ""
},
{
"first": "Anna",
"middle": [],
"last": "Currey",
"suffix": ""
},
{
"first": "Vishrav",
"middle": [],
"last": "Chaudhary",
"suffix": ""
},
{
"first": "Ivan Vladimir Meza",
"middle": [],
"last": "Ruiz",
"suffix": ""
},
{
"first": "Rolando",
"middle": [],
"last": "Coto-Solano",
"suffix": ""
},
{
"first": "Alexis",
"middle": [],
"last": "Palmer",
"suffix": ""
},
{
"first": "Elisabeth",
"middle": [],
"last": "Mager",
"suffix": ""
},
{
"first": "Ngoc",
"middle": [
"Thang"
],
"last": "Vu",
"suffix": ""
},
{
"first": "Graham",
"middle": [],
"last": "Neubig",
"suffix": ""
},
{
"first": "Katharina",
"middle": [],
"last": "Kann",
"suffix": ""
}
],
"year": 2021,
"venue": "Proceedings of the First Workshop on NLP for Indigenous Languages of the Americas",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Manuel Mager, Arturo Oncevay, Abteen Ebrahimi, John Ortega, Annette Rios, Angela Fan, Xi- mena Gutierrez-Vasques, Luis Chiruzzo, Gustavo Gim\u00e9nez-Lugo, Ricardo Ramos, Anna Currey, Vishrav Chaudhary, Ivan Vladimir Meza Ruiz, Rolando Coto-Solano, Alexis Palmer, Elisabeth Mager, Ngoc Thang Vu, Graham Neubig, and Katha- rina Kann. 2021. Findings of the AmericasNLP 2021 Shared Task on Open Machine Translation for Indigenous Languages of the Americas. In Proceed- ings of the First Workshop on NLP for Indigenous Languages of the Americas, Online. Association for Computational Linguistics.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "A\u00f1aani katonkosatzi parenini, El idioma del alto Peren\u00e9",
"authors": [
{
"first": "Elena",
"middle": [],
"last": "Mihas",
"suffix": ""
}
],
"year": 2011,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Elena Mihas. 2011. A\u00f1aani katonkosatzi parenini, El idioma del alto Peren\u00e9. Milwaukee, WI: Clarks Graphics.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Relatos huicholes = Wixarika' '\u00efxatsikayari",
"authors": [
{
"first": "Lucila",
"middle": [],
"last": "Mondrag\u00f3n",
"suffix": ""
},
{
"first": "Jacqueline",
"middle": [],
"last": "Tello",
"suffix": ""
},
{
"first": "Argelia",
"middle": [],
"last": "Valdez",
"suffix": ""
}
],
"year": 2002,
"venue": "CNCA-Direcci\u00f3n General de Culturas Populares e Ind\u00edgenas",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lucila Mondrag\u00f3n, Jacqueline Tello, and Argelia Valdez. 2002a. Relatos huicholes = Wixarika' '\u00efx- atsikayari. CNCA-Direcci\u00f3n General de Culturas Populares e Ind\u00edgenas, Ciudad de Mexico.",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Relatos otom\u00edes. Nfini H\u00f1\u00e4h\u00f1u. CNCA-Direcci\u00f3n General de Culturas Populares e Ind\u00edgenas",
"authors": [
{
"first": "Lucila",
"middle": [],
"last": "Mondrag\u00f3n",
"suffix": ""
},
{
"first": "Jacqueline",
"middle": [],
"last": "Tello",
"suffix": ""
},
{
"first": "Argelia",
"middle": [],
"last": "Valdez",
"suffix": ""
}
],
"year": 2002,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lucila Mondrag\u00f3n, Jacqueline Tello, and Argelia Valdez. 2002b. Relatos otom\u00edes. Nfini H\u00f1\u00e4h\u00f1u. CNCA-Direcci\u00f3n General de Culturas Populares e Ind\u00edgenas, Ciudad de Mexico.",
"links": null
},
"BIBREF27": {
"ref_id": "b27",
"title": "Universal Dependencies v2: An evergrowing multilingual treebank collection",
"authors": [
{
"first": "Joakim",
"middle": [],
"last": "Nivre",
"suffix": ""
},
{
"first": "Marie-Catherine",
"middle": [],
"last": "De Marneffe",
"suffix": ""
},
{
"first": "Filip",
"middle": [],
"last": "Ginter",
"suffix": ""
},
{
"first": "Jan",
"middle": [],
"last": "Haji\u010d",
"suffix": ""
},
{
"first": "Christopher",
"middle": [
"D"
],
"last": "Manning",
"suffix": ""
},
{
"first": "Sampo",
"middle": [],
"last": "Pyysalo",
"suffix": ""
},
{
"first": "Sebastian",
"middle": [],
"last": "Schuster",
"suffix": ""
},
{
"first": "Francis",
"middle": [],
"last": "Tyers",
"suffix": ""
},
{
"first": "Daniel",
"middle": [],
"last": "Zeman",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 12th Language Resources and Evaluation Conference",
"volume": "",
"issue": "",
"pages": "4034--4043",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Joakim Nivre, Marie-Catherine de Marneffe, Filip Gin- ter, Jan Haji\u010d, Christopher D. Manning, Sampo Pyysalo, Sebastian Schuster, Francis Tyers, and Daniel Zeman. 2020. Universal Dependencies v2: An evergrowing multilingual treebank collection. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 4034-4043, Mar- seille, France. European Language Resources Asso- ciation.",
"links": null
},
"BIBREF28": {
"ref_id": "b28",
"title": "Overcoming resistance: The normalization of an Amazonian tribal language",
"authors": [
{
"first": "John",
"middle": [],
"last": "Ortega",
"suffix": ""
},
{
"first": "Richard",
"middle": [
"Alexander"
],
"last": "Castro-Mamani",
"suffix": ""
},
{
"first": "Jaime Rafael Montoya",
"middle": [],
"last": "Samame",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 3rd Workshop on Technologies for MT of Low Resource Languages",
"volume": "",
"issue": "",
"pages": "1--13",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "John Ortega, Richard Alexander Castro-Mamani, and Jaime Rafael Montoya Samame. 2020. Overcom- ing resistance: The normalization of an Amazonian tribal language. In Proceedings of the 3rd Workshop on Technologies for MT of Low Resource Languages, pages 1-13, Suzhou, China. Association for Compu- tational Linguistics.",
"links": null
},
"BIBREF29": {
"ref_id": "b29",
"title": "A monolingual approach to contextualized word embeddings for mid-resource languages",
"authors": [
{
"first": "Pedro Javier Ortiz",
"middle": [],
"last": "Su\u00e1rez",
"suffix": ""
},
{
"first": "Laurent",
"middle": [],
"last": "Romary",
"suffix": ""
},
{
"first": "Beno\u00eet",
"middle": [],
"last": "Sagot",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
"volume": "",
"issue": "",
"pages": "1703--1714",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Pedro Javier Ortiz Su\u00e1rez, Laurent Romary, and Beno\u00eet Sagot. 2020. A monolingual approach to contextual- ized word embeddings for mid-resource languages. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1703-1714, Online. Association for Computational Linguistics.",
"links": null
},
"BIBREF30": {
"ref_id": "b30",
"title": "Parallel Global Voices: a collection of multilingual corpora with citizen media stories",
"authors": [
{
"first": "Prokopis",
"middle": [],
"last": "Prokopidis",
"suffix": ""
},
{
"first": "Vassilis",
"middle": [],
"last": "Papavassiliou",
"suffix": ""
},
{
"first": "Stelios",
"middle": [],
"last": "Piperidis",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16)",
"volume": "",
"issue": "",
"pages": "900--905",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Prokopis Prokopidis, Vassilis Papavassiliou, and Ste- lios Piperidis. 2016. Parallel Global Voices: a col- lection of multilingual corpora with citizen media stories. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16), pages 900-905, Portoro\u017e, Slovenia. Eu- ropean Language Resources Association (ELRA).",
"links": null
},
"BIBREF31": {
"ref_id": "b31",
"title": "Stanza: A Python natural language processing toolkit for many human languages",
"authors": [
{
"first": "Peng",
"middle": [],
"last": "Qi",
"suffix": ""
},
{
"first": "Yuhao",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "Yuhui",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "Jason",
"middle": [],
"last": "Bolton",
"suffix": ""
},
{
"first": "Christopher",
"middle": [
"D"
],
"last": "Manning",
"suffix": ""
}
],
"year": 2020,
"venue": "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Peng Qi, Yuhao Zhang, Yuhui Zhang, Jason Bolton, and Christopher D. Manning. 2020. Stanza: A Python natural language processing toolkit for many human languages. In Proceedings of the 58th An- nual Meeting of the Association for Computational Linguistics: System Demonstrations.",
"links": null
},
"BIBREF32": {
"ref_id": "b32",
"title": "Leveraging pre-trained checkpoints for sequence generation tasks",
"authors": [
{
"first": "Sascha",
"middle": [],
"last": "Rothe",
"suffix": ""
},
{
"first": "Shashi",
"middle": [],
"last": "Narayan",
"suffix": ""
},
{
"first": "Aliaksei",
"middle": [],
"last": "Severyn",
"suffix": ""
}
],
"year": 2020,
"venue": "Transactions of the Association for Computational Linguistics",
"volume": "8",
"issue": "",
"pages": "264--280",
"other_ids": {
"DOI": [
"10.1162/tacl_a_00313"
]
},
"num": null,
"urls": [],
"raw_text": "Sascha Rothe, Shashi Narayan, and Aliaksei Severyn. 2020. Leveraging pre-trained checkpoints for se- quence generation tasks. Transactions of the Asso- ciation for Computational Linguistics, 8:264-280.",
"links": null
},
"BIBREF33": {
"ref_id": "b33",
"title": "Quyllur Llaqtayuq Wawamanta",
"authors": [
{
"first": "Antoine",
"middle": [],
"last": "De Saint-Exup\u00e9ry",
"suffix": ""
}
],
"year": 2018,
"venue": "Ediciones El Lector",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Antoine de Saint-Exup\u00e9ry. 2018. Quyllur Llaqtayuq Wawamanta. Ediciones El Lector, Arequipa, Peru. Translated by Lydia Cornejo Endara & C\u00e9sar Itier.",
"links": null
},
"BIBREF34": {
"ref_id": "b34",
"title": "Improving neural machine translation models with monolingual data",
"authors": [
{
"first": "Rico",
"middle": [],
"last": "Sennrich",
"suffix": ""
},
{
"first": "Barry",
"middle": [],
"last": "Haddow",
"suffix": ""
},
{
"first": "Alexandra",
"middle": [],
"last": "Birch",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics",
"volume": "1",
"issue": "",
"pages": "86--96",
"other_ids": {
"DOI": [
"10.18653/v1/P16-1009"
]
},
"num": null,
"urls": [],
"raw_text": "Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016. Improving neural machine translation mod- els with monolingual data. In Proceedings of the 54th Annual Meeting of the Association for Compu- tational Linguistics (Volume 1: Long Papers), pages 86-96, Berlin, Germany. Association for Computa- tional Linguistics.",
"links": null
},
"BIBREF35": {
"ref_id": "b35",
"title": "Parallel data, tools and interfaces in OPUS",
"authors": [
{
"first": "J\u00f6rg",
"middle": [],
"last": "Tiedemann",
"suffix": ""
}
],
"year": 2012,
"venue": "Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12)",
"volume": "",
"issue": "",
"pages": "2214--2218",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J\u00f6rg Tiedemann. 2012. Parallel data, tools and inter- faces in OPUS. In Proceedings of the Eighth In- ternational Conference on Language Resources and Evaluation (LREC'12), pages 2214-2218, Istanbul, Turkey. European Language Resources Association (ELRA).",
"links": null
},
"BIBREF36": {
"ref_id": "b36",
"title": "Historia y alcances de un cuento para ni\u00f1os creado en el museo",
"authors": [
{
"first": "Felipe",
"middle": [],
"last": "Luis",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Villacorta",
"suffix": ""
}
],
"year": 2007,
"venue": "Illapa Mana Tukukuq",
"volume": "",
"issue": "4",
"pages": "101--112",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Luis Felipe Villacorta. 2007. \"Hab\u00eda una vez... El Per\u00fa de Antonio Raimondi\". Historia y alcances de un cuento para ni\u00f1os creado en el museo. Illapa Mana Tukukuq, (4):101-112.",
"links": null
}
},
"ref_entries": {
"TABREF1": {
"type_str": "table",
"num": null,
"text": "",
"html": null,
"content": "<table/>"
},
"TABREF2": {
"type_str": "table",
"num": null,
"text": "TEST Helsinki-2 (best) 1 .",
"html": null,
"content": "<table><tr><td>Set</td><td>System</td><td>Track</td><td/><td/><td/><td/><td colspan=\"2\">Languages</td><td/></tr><tr><td/><td/><td/><td>AYM</td><td>BZD</td><td>CNI</td><td>GN</td><td>HCH</td><td>NAH</td><td>OTO</td><td>QUY</td><td>SHP</td><td>TAR</td></tr><tr><td>DEV</td><td colspan=\"2\">CoAStaL-1: + extra data 1</td><td>.188</td><td>-</td><td colspan=\"2\">-.242</td><td colspan=\"2\">-.216</td><td colspan=\"2\">-.250</td><td>-</td><td>-</td></tr><tr><td/><td>CoAStaL-2: Random</td><td>2</td><td colspan=\"8\">.168 .107 .212 .128 .191 .184 .101 .232 .173 .113</td></tr><tr><td/><td>Baseline</td><td>2</td><td colspan=\"8\">.157 .068 .102 .193 .126 .157 .054 .304 .121 .039</td></tr><tr><td/><td/><td/><td/><td colspan=\"2\">(a) chrF</td><td/><td/><td/><td/></tr><tr><td/><td/><td/><td>AYM</td><td>BZD</td><td>CNI</td><td>GN</td><td>HCH</td><td>NAH</td><td>OTO</td><td>QUY</td><td>SHP</td><td>TAR</td></tr><tr><td>DEV</td><td colspan=\"2\">CoAStaL-1: Phrase-based 1 CoAStaL-2: Random 2</td><td colspan=\"4\">2.57 3.83 2.79 2.59 0.02 0.03 0.04 0.02</td><td colspan=\"4\">6.81 2.33 1.44 1.73 1.14 0.02 0.02 0.02</td><td>3.70 1.26 0.06 0.02</td></tr><tr><td/><td>Helsinki-2 (best)</td><td>1</td><td colspan=\"8\">2.80 5.18 6.09 8.92 15.67 3.25 5.59 5.38 10.49 3.56</td></tr><tr><td/><td colspan=\"2\">CoAStaL-1: Phrase-based 1</td><td colspan=\"4\">1.11 3.60 3.02 2.20</td><td colspan=\"4\">8.80 2.06 2.72 1.63</td><td>3.90 1.05</td></tr><tr><td>TEST</td><td colspan=\"2\">+ extra data 1</td><td>1.07</td><td>-</td><td colspan=\"2\">-2.24</td><td colspan=\"2\">-2.06</td><td colspan=\"2\">-1.24</td><td>-</td><td>-</td></tr><tr><td/><td>CoAStaL-2: Random</td><td>2</td><td colspan=\"4\">0.05 0.06 0.03 0.03</td><td colspan=\"4\">2.07 0.03 0.03 0.02</td><td>0.04 0.06</td></tr><tr><td/><td>Baseline</td><td>2</td><td colspan=\"4\">0.01 0.01 0.01 0.12</td><td colspan=\"4\">2.20 0.01 0.00 0.05</td><td>0.01 0.00</td></tr><tr><td/><td/><td/><td/><td colspan=\"2\">(b) BLEU</td><td/><td/><td/><td/></tr></table>"
}
}
}
} |