The instruction-tolerant sibling of DALabCommunity/Haidass-Translate-143M: same 143M zh⇄en translation training, plus 9.1% cleaned general-domain data (STEP_FUN ShareGPT) mixed in. Translation scores are within 0.3 BLEU of the pure-translation version, and the model retains limited general instruction-following ability that the pure-translation version does not have. Live demo: Haidass Translate Instruction on Hugging Face Spaces. *Drafter-143M: a control model with identical configuration, data and training recipe, except that it starts from random initialization instead of the pretrained base — used to quantify the contribution of base-model pretraining. OPUS-MT models are single-directional — one independent 78M model per direction; «-» marks directions a model does not serve. The same models re-evaluated on FLORES+ devtest (released 2026; zero overlap with dev): Mixing 9.1% general-domain data into the translation SFT gives this variant a limited ability to respond to general instructions (short poems, simple summaries, paraphrases) that the pure-translation sibling essentially lacks. This ability is fragile: prompts outside the translation domain may produce echoes of the…
Модальности:
Генерация текста
Области применения:
Диалог / чат Перевод Следование инструкциям
Задача: Генерация текста
Автор: umeiko
Теги: qwen3, translation, machine-translation, small-language-model, conversational, en, zh
Лайков: 3 | Загрузок: 157
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
