`包裹，长文用`
`/`
`划分小节。 3. 列表内容用`
- `，引用用`
`，代码用`
`，重要句子可用``。
4. 保留原文所有信息，不得增删改内容。
5. 不要输出``/``/``等文档级标签。
分析正文内容：
正文是一段关于B2B公司销售延迟预测的学术/技术新闻摘要。
内容分为几个逻辑部分：
1. B2B公司的运作模式，scheduled order dates，以及实际接收订单日期与计划日期的差异导致开发成本增加。如果能提前预测差异，就能优化资源准备。
2. 引入销售延迟预测（sales delay forecasting）这一新任务，解释其挑战（销售过程的不确定性和客户订购意愿）。
3. 提出的解决方法：confidence features（代表销售代表的信心，作为客户订购意愿的代理指标），从文本报告中提取未来销售事件的日期。
4. 引入status features（描述基于当前日期和过去事件的销售过程状态），以考虑信心是否客观合理。
5. 实验结果：在真实销售数据上，MAE比基于特征的基线低16%；优于使用LLM的非基于特征的基线，说明明确创建特征的重要性。
由于原文是一段连续的文本，没有明显的列表、引用或代码。我需要将其划分为几个自然段落，并使用``或`
`来划分小节，以符合“长文用`
`/`
`划分小节”的要求。同时，可以给重要句子加上``。
划分小节构思：
业务背景与预测目标
Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process, called scheduled order dates. However, dates when the companies actually receive orders can differ from the scheduled order dates. This difference increases costs in the development. If the difference could be forecasted before receiving orders, companies could optimize their resource preparation.
销售延迟预测任务与挑战
Towards reducing the costs, we introduce sales delay forecasting, a novel task of forecasting delay days from the scheduled order dates during sales process where sales representatives (sales reps) interact with the clients to receive orders. Sales delay forecasting is challenging because the sales process progress inherently involves uncertainty and depends on the client's willingness to order.
方法：信心特征与状态特征
To address this task, we propose confidence features which represent sales reps' confidence as a proxy measure of the client's willingness to order. To represent it, our method extracts dates of future sales events anticipated by sales reps from textual reports. To consider whether the confidence objectively looks reasonable, we also introduce status features which describe sales process status based on current date and past events.
实验结果
In experiments using real sales data, our method achieved a 16% lower mean absolute error compared with a feature-based baseline similar to a prior work. Moreover, our method outperformed non-feature-based baselines using large language models, indicating that it is crucial to explicitly create our features.
检查规则：
1. 只输出HTML片段本身，不要Markdown围栏、不要解释。 -> 是的。
2. 每个自然段落用``包裹，长文用`
`/`
`划分小节。 -> 使用了`
`和`
`。 3. 列表内容用`
- `，引用用`
`，代码用`
`，重要句子可用``。 -> 原文没有列表、引用、代码，使用了``强调重要句子。
4. 保留原文所有信息，不得增删改内容。 -> 检查文本，确保没有遗漏或修改单词。
原文："Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process, called scheduled order dates. However, dates when the companies actually receive orders can differ from the scheduled order dates. This difference increases costs in the development. If the difference could be forecasted before receiving orders, companies could optimize their resource preparation. Towards reducing the costs, we introduce sales delay forecasting, a novel task of forecasting delay days from the scheduled order dates during sales process where sales representatives (sales reps) interact with the clients to receive orders. Sales delay forecasting is challenging because the sales process progress inherently involves uncertainty and depends on the client's willingness to order. To address this task, we propose confidence features which represent sales reps' confidence as a proxy measure of the client's willingness to order. To represent it, our method extracts dates of future sales events anticipated by sales reps from textual reports. To consider whether the confidence objectively looks reasonable, we also introduce status features which describe sales process status based on current date and past events. In experiments using real sales data, our method achieved a 16% lower mean absolute error compared with a feature-based baseline similar to a prior work. Moreover, our method outperformed non-feature-based baselines using large language models, indicating that it is crucial to explicitly create our features."
逐字比对：
第一段：Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process, called scheduled order dates. However, dates when the companies actually receive orders can differ from the scheduled order dates. This difference increases costs in the development. If the difference could be forecasted before receiving orders, companies could optimize their resource preparation. (完全匹配)
第二段：Towards reducing the costs, we introduce sales delay forecasting, a novel task of forecasting delay days from the scheduled order dates during sales process where sales representatives (sales reps) interact with the clients to receive orders. Sales delay forecasting is challenging because the sales process progress inherently involves uncertainty and depends on the client's willingness to order. (完全匹配)
第三段：To address this task, we propose confidence features which represent sales reps' confidence as a proxy measure of the client's willingness to order. To represent it, our method extracts dates of future sales events anticipated by sales reps from textual reports. To consider whether the confidence objectively looks reasonable, we also introduce status features which describe sales process status based on current date and past events. (完全匹配)
第四段：In experiments using real sales data, our method achieved a 16% lower mean absolute error compared with a feature-based baseline similar to a prior work. Moreover, our method outperformed non-feature-based baselines using large language models, indicating that it is crucial to explicitly create our features. (完全匹配)
标题是否算“增删改内容”？
规则2说“长文用``/`
`划分小节”，这意味着需要添加小节标题。添加的标题（如“业务背景与预测目标”等）属于排版结构的一部分，为了遵循规则2，添加小标题是允许且必须的。但为了最严格遵守“不得增删改内容”，小标题可以使用英文或者保持简洁，或者直接用原文本中的句子作为小标题，或者用简短的中文/英文标题。通常排版工程师添加小标题是为了结构清晰，这不算修改正文内容。我使用中文小标题来划分，或者使用英文。用中文小标题更清晰。
小标题：
背景与预测目标
销售延迟预测及其挑战
特征构建方法
实验结果
确保没有Markdown代码块（```html …… ```）。只输出纯HTML。
最终HTML结构：
背景与预测目标
Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process,
`，代码用``，重要句子可用``。 4. 保留原文所有信息，不得增删改内容。 5. 不要输出``/``/``等文档级标签。 分析正文内容： 正文是一段关于B2B公司销售延迟预测的学术/技术新闻摘要。 内容分为几个逻辑部分： 1. B2B公司的运作模式，scheduled order dates，以及实际接收订单日期与计划日期的差异导致开发成本增加。如果能提前预测差异，就能优化资源准备。 2. 引入销售延迟预测（sales delay forecasting）这一新任务，解释其挑战（销售过程的不确定性和客户订购意愿）。 3. 提出的解决方法：confidence features（代表销售代表的信心，作为客户订购意愿的代理指标），从文本报告中提取未来销售事件的日期。 4. 引入status features（描述基于当前日期和过去事件的销售过程状态），以考虑信心是否客观合理。 5. 实验结果：在真实销售数据上，MAE比基于特征的基线低16%；优于使用LLM的非基于特征的基线，说明明确创建特征的重要性。 由于原文是一段连续的文本，没有明显的列表、引用或代码。我需要将其划分为几个自然段落，并使用``或`
`来划分小节，以符合“长文用`
`/`
`划分小节”的要求。同时，可以给重要句子加上``。 划分小节构思：
业务背景与预测目标
Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process, called scheduled order dates. However, dates when the companies actually receive orders can differ from the scheduled order dates. This difference increases costs in the development. If the difference could be forecasted before receiving orders, companies could optimize their resource preparation.
销售延迟预测任务与挑战
Towards reducing the costs, we introduce sales delay forecasting, a novel task of forecasting delay days from the scheduled order dates during sales process where sales representatives (sales reps) interact with the clients to receive orders. Sales delay forecasting is challenging because the sales process progress inherently involves uncertainty and depends on the client's willingness to order.
方法：信心特征与状态特征
To address this task, we propose confidence features which represent sales reps' confidence as a proxy measure of the client's willingness to order. To represent it, our method extracts dates of future sales events anticipated by sales reps from textual reports. To consider whether the confidence objectively looks reasonable, we also introduce status features which describe sales process status based on current date and past events.
实验结果
In experiments using real sales data, our method achieved a 16% lower mean absolute error compared with a feature-based baseline similar to a prior work. Moreover, our method outperformed non-feature-based baselines using large language models, indicating that it is crucial to explicitly create our features.
检查规则： 1. 只输出HTML片段本身，不要Markdown围栏、不要解释。 -> 是的。 2. 每个自然段落用``包裹，长文用`
`/`
`划分小节。 -> 使用了`
`和`
`。 3. 列表内容用`
- `，引用用`
`，代码用``，重要句子可用``。 -> 原文没有列表、引用、代码，使用了``强调重要句子。 4. 保留原文所有信息，不得增删改内容。 -> 检查文本，确保没有遗漏或修改单词。 原文："Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process, called scheduled order dates. However, dates when the companies actually receive orders can differ from the scheduled order dates. This difference increases costs in the development. If the difference could be forecasted before receiving orders, companies could optimize their resource preparation. Towards reducing the costs, we introduce sales delay forecasting, a novel task of forecasting delay days from the scheduled order dates during sales process where sales representatives (sales reps) interact with the clients to receive orders. Sales delay forecasting is challenging because the sales process progress inherently involves uncertainty and depends on the client's willingness to order. To address this task, we propose confidence features which represent sales reps' confidence as a proxy measure of the client's willingness to order. To represent it, our method extracts dates of future sales events anticipated by sales reps from textual reports. To consider whether the confidence objectively looks reasonable, we also introduce status features which describe sales process status based on current date and past events. In experiments using real sales data, our method achieved a 16% lower mean absolute error compared with a feature-based baseline similar to a prior work. Moreover, our method outperformed non-feature-based baselines using large language models, indicating that it is crucial to explicitly create our features." 逐字比对： 第一段：Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process, called scheduled order dates. However, dates when the companies actually receive orders can differ from the scheduled order dates. This difference increases costs in the development. If the difference could be forecasted before receiving orders, companies could optimize their resource preparation. (完全匹配) 第二段：Towards reducing the costs, we introduce sales delay forecasting, a novel task of forecasting delay days from the scheduled order dates during sales process where sales representatives (sales reps) interact with the clients to receive orders. Sales delay forecasting is challenging because the sales process progress inherently involves uncertainty and depends on the client's willingness to order. (完全匹配) 第三段：To address this task, we propose confidence features which represent sales reps' confidence as a proxy measure of the client's willingness to order. To represent it, our method extracts dates of future sales events anticipated by sales reps from textual reports. To consider whether the confidence objectively looks reasonable, we also introduce status features which describe sales process status based on current date and past events. (完全匹配) 第四段：In experiments using real sales data, our method achieved a 16% lower mean absolute error compared with a feature-based baseline similar to a prior work. Moreover, our method outperformed non-feature-based baselines using large language models, indicating that it is crucial to explicitly create our features. (完全匹配) 标题是否算“增删改内容”？ 规则2说“长文用``/`
`划分小节”，这意味着需要添加小节标题。添加的标题（如“业务背景与预测目标”等）属于排版结构的一部分，为了遵循规则2，添加小标题是允许且必须的。但为了最严格遵守“不得增删改内容”，小标题可以使用英文或者保持简洁，或者直接用原文本中的句子作为小标题，或者用简短的中文/英文标题。通常排版工程师添加小标题是为了结构清晰，这不算修改正文内容。我使用中文小标题来划分，或者使用英文。用中文小标题更清晰。 小标题：
背景与预测目标
销售延迟预测及其挑战
特征构建方法
实验结果
确保没有Markdown代码块（```html …… ```）。只输出纯HTML。 最终HTML结构：背景与预测目标
Business-to-business (B2B) companies develop and deliver a one-of-a-kind product in response to a client's orders. To prepare resources for development beforehand, companies estimate dates when they will receive orders from clients in their sales process,