` 包裹。使用 `
` 或 `
` 进行章节划分。 * *规则 3：* 列表使用 `
- `，引用使用 `
`，代码使用 `
`，重要句子使用 ``。
* *规则 4：* 保留原文所有信息，不得增删改内容。
* *规则 5：* 不输出 ``、`` 或 ``。
4. 详细文本拆解与 HTML 映射：
* *原文第 1 段：* "EmbeddingGemma-300M is a small open embedder whose tokenizer represents every Tajik letter, and as released it is weak at Tajik retrieval: on our Tajik law test the right passage is in its top ten results for 49.3% of questions, on tax documents for 45.9%, and on the human-written Belebele Tajik set for 69.8%, against 98.6% on the Russian version of the same set."
* *映射：* 这是一个关于原始模型的非常长的段落。为了提高可读性，可以将其拆分或直接保留为一个段落，并加上 `` 标签。由于第 2 段引入了一个新的模型，可以将其作为引言段落。对模型名称使用 ``。
* `
EmbeddingGemma-300M is a small open embedder whose tokenizer represents every Tajik letter, and as released it is weak at Tajik retrieval: on our Tajik law test the right passage is in its top ten results for 49.3% of questions, on tax documents for 45.9%, and on the human-written Belebele Tajik set for 69.8%, against 98.6% on the Russian version of the same set.
`
* *原文第 2 段：* "We present Saidzoda-Tajik-EmbeddingGemma-300M, the same model fully fine-tuned for one epoch on 235,140 question and passage pairs written by our own Tajik language model for laws, tax documents, Wikipedia and web text, with the same mined hard negatives, English and Russian replay, loss, global batch, learning rate and epoch as our earlier Saidzoda-Tajik-E5-Large (arXiv:2611.00592) and BGE-M3 models."
* *映射：* 引入了新模型和训练细节。对新模型名称和 arXiv 代码（或许可以去掉括号，或者保留文本原样）使用 ``。arXiv 代码实际上是文本中的一个超链接，但要求是“保留所有信息，不得增删改”。为了安全起见并严格遵守规则 4（不得增删改内容），保持纯文本，并使用 `` 来标注索引号？不，原文是 `(arXiv:2611.00592)`。我将严格按照文本进行包装。使用 `` 标签。 * `
We present Saidzoda-Tajik-EmbeddingGemma-300M, the same model fully fine-tuned for one epoch on 235,140 question and passage pairs written by our own Tajik language model for laws, tax documents, Wikipedia and web text, with the same mined hard negatives, English and Russian replay, loss, global batch, learning rate and epoch as our earlier Saidzoda-Tajik-E5-Large (arXiv:2611.00592) and BGE-M3 models.
`
* *原文第 3 段：* "After training, law reaches 92.4% and tax 91.7%; on Belebele, the one test written by people, recall@10 rises from 69.8% to 92.6% and the first result is right for 76.6% of questions instead of 51.7%."
* *映射：* 性能结果。使用 `` 标签处理。对关键指标使用 ``。 * `
After training, law reaches 92.4% and tax 91.7%; on Belebele, the one test written by people, recall@10 rises from 69.8% to 92.6% and the first result is right for 76.6% of questions instead of 51.7%.
` * *原文第 4 段：* "A checkpoint at step 100 of 466 already holds 88% of the law gain at recall@10." * *映射：* 单独的事实。使用 `
` 标签处理。 * `
A checkpoint at step 100 of 466 already holds 88% of the law gain at recall@10.
` * *原文第 5 段：* "At recall@10 the result is no more than 2.8 points below the two 560M models trained on the same pairs on law, tax and Wikipedia, within the intervals, with 768-dimensional vectors and about half their parameters; at recall@1, for which we have no intervals, it stays 7.0 points behind the E5 model on Belebele." * *映射：* 模型对比。使用 `
` 标签处理。对关键数字使用 ``。 * `
At recall@10 the result is no more than 2.8 points below the two 560M models trained on the same pairs on law, tax and Wikipedia, within the intervals, with 768-dimensional vectors and about half their parameters; at recall@1, for which we have no intervals, it stays 7.0 points behind the E5 model on Belebele.
` * *原文第 6 段：* "The cost is a drop of 3.0 and 2.8 points in Russian and English recall@1, larger than for the two earlier models." * *映射：* 局限性/负面影响。使用 `
` 标签处理。 * `
The cost is a drop of 3.0 and 2.8 points in Russian and English recall@1, larger than for the two earlier models.
` * *原文第 7 段：* "We also report that, according to our run notes, a bfloat16 test run produced a non-finite loss at step 3 which the training framework logged as zero, so training was done in fp32." * *映射：* 技术说明。这可以被视为一个独立的章节，或者作为一个段落。文本中提到了“`bfloat16`”和“`fp32`”。为了使其看起来更像代码/技术文本，可以使用 `
`，代码使用 ``，重要句子使用 ``。 * *规则 4：* 保留原文所有信息，不得增删改内容。 * *规则 5：* 不输出 ``、`` 或 ``。 4. 详细文本拆解与 HTML 映射： * *原文第 1 段：* "EmbeddingGemma-300M is a small open embedder whose tokenizer represents every Tajik letter, and as released it is weak at Tajik retrieval: on our Tajik law test the right passage is in its top ten results for 49.3% of questions, on tax documents for 45.9%, and on the human-written Belebele Tajik set for 69.8%, against 98.6% on the Russian version of the same set." * *映射：* 这是一个关于原始模型的非常长的段落。为了提高可读性，可以将其拆分或直接保留为一个段落，并加上 `` 标签。由于第 2 段引入了一个新的模型，可以将其作为引言段落。对模型名称使用 ``。 * `
EmbeddingGemma-300M is a small open embedder whose tokenizer represents every Tajik letter, and as released it is weak at Tajik retrieval: on our Tajik law test the right passage is in its top ten results for 49.3% of questions, on tax documents for 45.9%, and on the human-written Belebele Tajik set for 69.8%, against 98.6% on the Russian version of the same set.
` * *原文第 2 段：* "We present Saidzoda-Tajik-EmbeddingGemma-300M, the same model fully fine-tuned for one epoch on 235,140 question and passage pairs written by our own Tajik language model for laws, tax documents, Wikipedia and web text, with the same mined hard negatives, English and Russian replay, loss, global batch, learning rate and epoch as our earlier Saidzoda-Tajik-E5-Large (arXiv:2611.00592) and BGE-M3 models." * *映射：* 引入了新模型和训练细节。对新模型名称和 arXiv 代码（或许可以去掉括号，或者保留文本原样）使用 ``。arXiv 代码实际上是文本中的一个超链接，但要求是“保留所有信息，不得增删改”。为了安全起见并严格遵守规则 4（不得增删改内容），保持纯文本，并使用 `` 来标注索引号？不，原文是 `(arXiv:2611.00592)`。我将严格按照文本进行包装。使用 `` 标签。 * `
We present Saidzoda-Tajik-EmbeddingGemma-300M, the same model fully fine-tuned for one epoch on 235,140 question and passage pairs written by our own Tajik language model for laws, tax documents, Wikipedia and web text, with the same mined hard negatives, English and Russian replay, loss, global batch, learning rate and epoch as our earlier Saidzoda-Tajik-E5-Large (arXiv:2611.00592) and BGE-M3 models.
` * *原文第 3 段：* "After training, law reaches 92.4% and tax 91.7%; on Belebele, the one test written by people, recall@10 rises from 69.8% to 92.6% and the first result is right for 76.6% of questions instead of 51.7%." * *映射：* 性能结果。使用 `` 标签处理。对关键指标使用 ``。 * `
After training, law reaches 92.4% and tax 91.7%; on Belebele, the one test written by people, recall@10 rises from 69.8% to 92.6% and the first result is right for 76.6% of questions instead of 51.7%.
` * *原文第 4 段：* "A checkpoint at step 100 of 466 already holds 88% of the law gain at recall@10." * *映射：* 单独的事实。使用 `` 标签处理。 * `
A checkpoint at step 100 of 466 already holds 88% of the law gain at recall@10.
` * *原文第 5 段：* "At recall@10 the result is no more than 2.8 points below the two 560M models trained on the same pairs on law, tax and Wikipedia, within the intervals, with 768-dimensional vectors and about half their parameters; at recall@1, for which we have no intervals, it stays 7.0 points behind the E5 model on Belebele." * *映射：* 模型对比。使用 `` 标签处理。对关键数字使用 ``。 * `
At recall@10 the result is no more than 2.8 points below the two 560M models trained on the same pairs on law, tax and Wikipedia, within the intervals, with 768-dimensional vectors and about half their parameters; at recall@1, for which we have no intervals, it stays 7.0 points behind the E5 model on Belebele.
` * *原文第 6 段：* "The cost is a drop of 3.0 and 2.8 points in Russian and English recall@1, larger than for the two earlier models." * *映射：* 局限性/负面影响。使用 `` 标签处理。 * `
The cost is a drop of 3.0 and 2.8 points in Russian and English recall@1, larger than for the two earlier models.
` * *原文第 7 段：* "We also report that, according to our run notes, a bfloat16 test run produced a non-finite loss at step 3 which the training framework logged as zero, so training was done in fp32." * *映射：* 技术说明。这可以被视为一个独立的章节，或者作为一个段落。文本中提到了“`bfloat16`”和“`fp32`”。为了使其看起来更像代码/技术文本，可以使用 `