{ "input": "/projects/prudlab/formosan_parallel_corpora/splits_zh_v1/big_corpus_zh_in_domain_hard.csv", "model": "/scratch/scheppat/formosan_mt_experiments/runs/E3_zh_spm8192_f2zh_zh_e3_20260510-155741/final", "tokenizer": "/scratch/scheppat/formosan_mt_experiments/runs/E3_zh_spm8192_f2zh_zh_e3_20260510-155741/final", "direction": "f2zh", "target_lang": "chinese", "bleu_tokenize": "zh", "split": "test", "samples": 37435, "global": { "samples": 37435, "BLEU": 9.788873522245254, "chrF2": 11.773185537762172, "TER": 109.28997211806558 }, "by_language": { "ami": { "samples": 4866, "BLEU": 9.67664596964354, "chrF2": 11.257236746158707, "TER": 107.56081525312295 }, "bnn": { "samples": 3223, "BLEU": 9.438797365141522, "chrF2": 10.522835474870561, "TER": 105.0420168067227 }, "ckv": { "samples": 1832, "BLEU": 14.267338682634806, "chrF2": 15.110294265244454, "TER": 108.67237687366168 }, "dru": { "samples": 3846, "BLEU": 5.572741778754694, "chrF2": 8.574754937863247, "TER": 108.67220849247362 }, "pwn": { "samples": 3221, "BLEU": 7.858502138390441, "chrF2": 9.953520893642228, "TER": 106.32911392405062 }, "pyu": { "samples": 2225, "BLEU": 15.76590468147179, "chrF2": 16.92931971633205, "TER": 124.54545454545453 }, "ssf": { "samples": 1180, "BLEU": 14.731852581704961, "chrF2": 16.35588870724647, "TER": 106.68789808917198 }, "sxr": { "samples": 1106, "BLEU": 9.174525069515981, "chrF2": 12.793965259986544, "TER": 108.06591500433652 }, "szy": { "samples": 1601, "BLEU": 11.415668309044698, "chrF2": 14.82761299370274, "TER": 104.58435207823962 }, "tao": { "samples": 1455, "BLEU": 6.925709253385289, "chrF2": 10.383676109942837, "TER": 110.84583901773533 }, "tay": { "samples": 4293, "BLEU": 6.83184841971991, "chrF2": 9.06833672004013, "TER": 108.88795282224095 }, "trv": { "samples": 4573, "BLEU": 10.091903008657704, "chrF2": 11.960338890715946, "TER": 109.35416666666666 }, "tsu": { "samples": 1250, "BLEU": 8.564382999347023, "chrF2": 11.415204943663671, "TER": 126.83504340962905 }, "xnb": { "samples": 1552, "BLEU": 12.963985306907468, "chrF2": 15.187610721437808, "TER": 107.91100123609394 }, "xsy": { "samples": 1212, "BLEU": 14.445934092992472, "chrF2": 16.31696596414603, "TER": 106.79611650485437 } }, "by_source_bucket": { "Formosan-100_Paiwan_Texts": { "samples": 422, "BLEU": 7.6358528438610955, "chrF2": 8.786253362904043, "TER": 101.16279069767442 }, "Formosan-Amis_myths_and_customs": { "samples": 134, "BLEU": 1.3834967492612475, "chrF2": 3.592488073042097, "TER": 104.22993492407808 }, "Formosan-Old_Texts": { "samples": 8, "BLEU": 1.4198641212108538, "chrF2": 4.022994803599383, "TER": 100.0 }, "Formosan-PaiwanStories": { "samples": 25, "BLEU": 9.577235523415832, "chrF2": 10.935197468097126, "TER": 100.0 }, "Formosan-Rik-Bunun": { "samples": 225, "BLEU": 1.6136362126845787, "chrF2": 3.5441301624591186, "TER": 104.88721804511279 }, "Formosan-ePark": { "samples": 241, "BLEU": 48.68162932289159, "chrF2": 44.69693315027365, "TER": 92.27642276422763 }, "Formosan-gitbook_translations": { "samples": 93, "BLEU": 3.9094128957776606, "chrF2": 5.7929680228277745, "TER": 100.0 }, "culture": { "samples": 9919, "BLEU": 6.549705149226013, "chrF2": 8.025035060319183, "TER": 106.47144948755492 }, "essays": { "samples": 818, "BLEU": 25.800187065494686, "chrF2": 26.181381000355476, "TER": 99.11209766925639 }, "ntu": { "samples": 1950, "BLEU": 10.931310727633916, "chrF2": 11.96563503066236, "TER": 102.64423076923077 }, "picture_book": { "samples": 17068, "BLEU": 11.985017998958577, "chrF2": 14.739335041727964, "TER": 112.57698727911128 }, "picture_story": { "samples": 558, "BLEU": 9.328673539386655, "chrF2": 10.251406596288156, "TER": 102.64900662251655 }, "presidential_apology": { "samples": 8, "BLEU": 3.2671889237498863, "chrF2": 6.082422547344087, "TER": 120.0 }, "reading_writing": { "samples": 3711, "BLEU": 7.548475476363713, "chrF2": 8.953578848891137, "TER": 109.46478873239435 }, "youtube": { "samples": 2255, "BLEU": 7.694036279346885, "chrF2": 9.527966515275404, "TER": 114.25576519916143 } }, "by_length_bin": { "004_008": { "samples": 720, "BLEU": 13.18000769800177, "chrF2": 15.45653407120694, "TER": 98.3225806451613 }, "009_016": { "samples": 12886, "BLEU": 11.737148849330296, "chrF2": 13.767594349322392, "TER": 106.43694493783305 }, "017_032": { "samples": 19581, "BLEU": 9.967727275396937, "chrF2": 11.846840916327807, "TER": 109.10390167214521 }, "033_plus": { "samples": 4248, "BLEU": 7.733152016372462, "chrF2": 10.009223680215971, "TER": 118.41055006073226 } } }