`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。
分析提供的正文：
"Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence. Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier. On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts."
这是一段学术摘要/新闻正文，没有明显的列表、引用或代码。可以将其划分为几个小节，使用``或`
`，或者直接用`
`包裹。因为内容不长，可以分成几个段落，或者加几个小标题来体现排版能力。不过规则说“长文用`
`，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。 分析提供的正文： "Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence. Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier. On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts." 这是一段学术摘要/新闻正文，没有明显的列表、引用或代码。可以将其划分为几个小节，使用``或`
`，或者直接用`
`包裹。因为内容不长，可以分成几个段落，或者加几个小标题来体现排版能力。不过规则说“长文用`
`/`
`划分小节”，这段不算特别长，但为了体现结构，可以加小标题，或者直接按句子组合成几个段落。 仔细看原文，其实是一整段。如果直接用一个`
`包裹，可能不符合“每个自然段落用`
`包裹，长文用`
`/`
`划分小节”的暗示。我可以把原文按逻辑划分为几个部分，并加上小标题，或者就按自然段处理。原文没有换行，算作一个自然段。但为了排版好看，我可以手动根据逻辑切分段落，但规则说“不得增删改内容”，切分段落不算改内容，只是排版。 或者，我可以直接把它作为一段，用`
`包裹，里面用``强调重要句子。 让我们尝试切分为几个逻辑段落，并加上小标题，这样更符合“排版工程师”的角色。 比如：
研究背景与方法
Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence.
核心机制
Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier.
实验结果与结论
On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts.
检查规则： 1. 只输出HTML片段本身，不要Markdown围栏、不要解释。 -> 满足。 2. 每个自然段落用``包裹，长文用`
`/`
`划分小节。 -> 满足。
3. 列表内容用`- `，引用用`
`，代码用`
`，重要句子可用``。 -> 没有列表、引用、代码，使用了``。
4. 保留原文所有信息，不得增删改内容。 -> 检查切分后的文本是否完全一致。
原文：
Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence. Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier. On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts.
切分1：
Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence.
切分2：
Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier.
切分3：
On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts.
完全一致。
小标题是新增的，规则说“不得增删改内容”，小标题算不算增删内容？
“长文用/
划分小节”，这意味着可以添加小标题来划分小节。如果担心小标题被算作“增删内容”，也可以不加小标题，直接用`
`分段，或者把整段作为一个`
`。
但“划分小节”通常意味着需要有小节标题。为了安全起见，我可以使用语义化的标签，比如直接分段，不加文字标题，或者加简单的标题。其实“划分小节”可以通过`
`，代码用``，重要句子可用``。 -> 没有列表、引用、代码，使用了``。 4. 保留原文所有信息，不得增删改内容。 -> 检查切分后的文本是否完全一致。 原文： Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence. Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier. On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts. 切分1： Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence. 切分2： Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier. 切分3： On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts. 完全一致。 小标题是新增的，规则说“不得增删改内容”，小标题算不算增删内容？ “长文用/
划分小节”，这意味着可以添加小标题来划分小节。如果担心小标题被算作“增删内容”，也可以不加小标题，直接用`
`分段，或者把整段作为一个`
`。 但“划分小节”通常意味着需要有小节标题。为了安全起见，我可以使用语义化的标签，比如直接分段，不加文字标题，或者加简单的标题。其实“划分小节”可以通过`
`/`
`来实现，如果原文没有标题，强行加标题可能违反“不得增删改内容”。 更安全的做法：不添加额外的文字作为标题，而是将原文合理分段，如果必须用`
`/`
`，可以使用原文中的关键句作为标题？不行，那会打乱原文。 或者，原文本身就不算“长文”，所以不需要用`
`/`
`划分小节，直接用`
`包裹自然段落即可。 原文是一整段。我可以把它分成几个自然