Dear all about the performance of smt-e/c on dialog: for more than 0.5 million training sentences, the result of dialog sentences is 5.5542 while the result of last year (by Neon) is 5.92, still a long way apart. but i managed to translate from japanese, english, french to chinese. earlier, i did experiments from chinese to english and french. to make things easier, i will try to make one exe capable of translation between all 6 pairs, at the expense of running a little slower. but speed does not matter for smt. still much work needs to be done. anyone willing to help correct corpus, please let me know Best Regards Professor Mandel Shi Department of Computer Science Xiamen University 361005, Xiamen Fujian, China Tel: +86-592-218-8355 2004-09-13