`包裹,长文用`
`/`
`划分小节(这里内容不长,可能不需要或者简单分一下,或者直接用`
`,因为只有一段。为了符合规则,如果只有一段,就只用`
`,或者加个`
`标题?用户说“长文用h2/h3划分小节”,这篇不算长文,可以不加标题,或者加一个适当的标题。不过原文没有标题,规则4说“保留原文所有信息,不得增删改内容”,所以我不能自己加标题。直接用`
`包裹即可。) 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。 分析原文内容: "Retrieval-augmented generation (RAG) improves factuality by grounding large language models (LLMs) in external evidence. However, long retrieved contexts incur computational overhead and can even hurt performance due to noisy or irrelevant content. We propose \textsc{StitchRAG}, a framework for constructing compact context in RAG. \textsc{StitchRAG} builds a faithful, information-dense context while dynamically compressing the retrieved passages. We adapt the size reduction ratio to each passage by formulating a utility maximization trading-off the retained information and computation costs. We then extract answer candidates and query-related keywords from passages as anchors and collect passage sentences containing anchors under the derived size constraint. Finally, we connect or stitch the collected sentences together using text diffusion to form a coherent and natural context. Experiments on five open-domain QA benchmarks show that \textsc{StitchRAG} consistently outperforms diverse multi-step and compression baselines. Extensive ablations show the effectiveness of dynamic size allocation, anchor preservation, and diffusion-based stitching." 原文是一整段。我可以将其分成几个自然段,或者保持为一个``。为了排版美观,可以根据语义拆分成几个`
`,但规则说“每个自然段落用
包裹”,原文只有一个自然段。不过把长句拆分成多个`
`可能算“增删改”?不,排版时按语义断句成多个`
`也是常见的,但最安全的是保持原文的一个自然段,用一个`
`包裹。 或者,我可以把其中的重点用``标出。比如"We propose \textsc{StitchRAG}, a framework for constructing compact context in RAG." 可以加粗。 关于`\textsc{StitchRAG}`,这是LaTeX的字体命令,在HTML中可以转换为`StitchRAG`或者直接保留文本`StitchRAG`,或者用`
StitchRAG`?通常排版时,小写大写字体可以用CSS,但纯HTML片段中可以用`StitchRAG`或者直接输出`StitchRAG`。为了不改变内容,直接输出`StitchRAG`或者用`StitchRAG`。更简单的是直接写`StitchRAG`,因为`\textsc`是排版命令,转成HTML的小号大写字体更合理。不过直接保留`StitchRAG`最稳妥,避免被认为增删。我用`StitchRAG`来体现排版工程师的专业性,或者直接用`StitchRAG`。考虑到规则4“不得增删改内容”,把`\textsc{StitchRAG}`转为`StitchRAG`是去掉了LaTeX标记,属于合理排版。 让我们把原文按逻辑分成几个``段落,以提升可读性(排版工程师的职责): 段落1:背景与问题。 Retrieval-augmented generation (RAG) improves factuality by grounding large language models (LLMs) in external evidence. However, long retrieved contexts incur computational overhead and can even hurt performance due to noisy or irrelevant content. 段落2:方法概述。 We propose StitchRAG, a framework for constructing compact context in RAG. StitchRAG builds a faithful, information-dense context while dynamically compressing the retrieved passages. 段落3:方法细节。 We adapt the size reduction ratio to each passage by formulating a utility maximization trading-off the retained information and computation costs. We then extract answer candidates and query-related keywords from passages as anchors and collect passage sentences containing anchors under the derived size constraint. 段落4:方法细节与实验结果。 Finally, we connect or stitch the collected sentences together using text diffusion to form a coherent and natural context. Experiments on five open-domain QA benchmarks show that StitchRAG consistently outperforms diverse multi-step and compression baselines. Extensive ablations show the effectiveness of dynamic size allocation, anchor preservation, and diffusion-based stitching. 检查规则: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 -> 好的。 2. 每个自然段落用
包裹,长文用
/
划分小节。 -> 没有长文,不用h2/h3。 3. 列表内容用
- ,引用用
,代码用
,重要句子可用。 -> 使用了。 4. 保留原文所有信息,不得增删改内容。 -> 拆分段落不改变文字内容。去掉`\textsc`属于格式转换。 5. 不要输出//等文档级标签。 -> 好的。 关于 `\textsc{StitchRAG}`,我可以用 `StitchRAG` 来完美还原 LaTeX 的 `\textsc` 效果,这体现了“专业新闻排版工程师”的素养。 构建HTML:Retrieval-augmented generation (RAG) improves factuality by grounding large language models (LLMs) in external evidence. However, long retrieved contexts incur computational overhead and can even hurt performance due to noisy or irrelevant content.
We propose StitchRAG, a framework for constructing compact context in RAG. StitchRAG builds a faithful, information-dense context while dynamically compressing the retrieved passages.
We adapt the size reduction ratio to each passage by formulating a utility maximization trading-off the retained information and computation costs. We then extract answer candidates and query-related keywords from passages as anchors and collect passage sentences containing anchors under the derived size constraint.
Finally, we connect or stitch the collected sentences together using text diffusion to form a coherent and natural context. Experiments on five open-domain QA benchmarks show that StitchRAG consistently outperforms diverse multi-step and compression baselines. Extensive ablations show the effectiveness of dynamic size allocation, anchor preservation, and diffusion-based stitching.
再审视一下规则4:“不得增删改内容”。把一段话拆成四段,算不算“改内容”?通常排版中按语义分段是允许的,但为了最严格地遵守“不得增删改内容”,也许保持原来的一个自然段,只用一个``包裹是最安全的。 原文: Retrieval-augmented generation (RAG) improves factuality by grounding large language models (LLMs) in external evidence. However, long retrieved contexts incur computational overhead and can even hurt performance due to noisy or irrelevant content. We propose \textsc{StitchRAG}, a framework for constructing compact context in RAG. \textsc{StitchRAG} builds a faithful, information-dense context while
- ,引用用