首页 > AI前沿 > Knowledge Transfer for Conditional Flow Matching: Frozen Estimation and Adaptation

Knowledge Transfer for Conditional Flow Matching: Frozen Estimation and Adaptation

arXiv机器学习 2026-11-11 15:43 21 阅读 查看原文
用户需要将一段纯文本的英文新闻正文,转换成语义化的HTML片段。 规则要求: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 2. 每个自然段落用`

`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:
    正文是一段关于条件流匹配迁移学习的学术摘要,包含多个句子,实际上是一个长段落,或者可以按逻辑划分为几个小节。
    原文:
    "Conditional generative modeling with limited target data can benefit from representations learned in a related source population. We develop a transfer learning framework for conditional flow matching that accommodates distinct source and target response distributions and dimensions. Two procedures exploit a representation learned from source data: TFM-Freeze estimates a target velocity head with the representation fixed, whereas TFM-Adapt jointly learns the head and a representation adaptation under a functional anchoring penalty. For explicit neural network classes, we establish bounds on conditional squared Wasserstein risk that account for source estimation, target fitting, representation mismatch, and numerical integration. When the source representation admits a smooth correction, adaptation replaces the frozen mismatch contribution by the cost of learning the correction and a controllable regularization bias. An information lower bound identifies regimes in which adaptation attains a strictly faster conditional squared Wasserstein risk rate than any generator restricted to the frozen context information. We also develop independent validation procedure for selecting among fitted conditional generators. We show that the selection procedure satisfies an oracle inequality, and the switching rule guarantees, with high probability, no increase in conditional energy risk relative to a target-only generator. Simulations and real data applications illustrate the gains from our proposed methods and the effectiveness of target validation in guarding against negative transfer."
    
    需要划分小节并使用`

    `或`

    `。由于这是一篇摘要,可以划分为几个主题部分,但必须保留原文所有信息,不得增删改内容。如果添加小标题,小标题算“增改内容”吗?规则说“不得增删改内容”,这意味着不能自己编造小标题文字,或者把原有的句子改写。但如果原文没有小标题,我直接用`

    `包裹整个段落也是可以的。不过规则2提到“长文用`

    `/`

    `划分小节”,如果不添加标题,怎么划分小节?可以把原有的句子作为小节标题?不行,那会改变原文结构。 或者,整个文本就是一个自然段落,直接用一个`

    `包裹。但规则说“长文用`

    `/`

    `划分小节”,如果不划分,可能不符合规则。 等等,我可以把第一句或者某些句子作为小节标题?不,“不得增删改内容”。如果我添加了HTML标签,不改变文字内容。 如果原文没有明显的小节,我是否应该将整个内容放在一个`

    `中,或者将某些句子拆分为多个`

    `,并使用``强调重要句子,而不需要强行加`

    `,因为加`

    `必然需要文字,而添加文字属于“增”内容。 仔细看规则:“保留原文所有信息,不得增删改内容”。如果我不添加任何额外的标题文字,只使用`

    `,这也是符合“保留原文所有信息,不得增删改”的。为了稳妥,我把整个文本作为一个或几个`

    `,不添加自创的标题,以免违反“不得增删改内容”的规则。或者,我可以把原文中的某些完整句子用`

    `包裹,重要句子用``。 原文其实是一个连续的段落。我可以将其拆分为几个`

    `,按逻辑断句,但文字一字不改。 例如: `

    Conditional generative modeling with limited target data can benefit from representations learned in a related source population. We develop a transfer learning framework for conditional flow matching that accommodates distinct source and target response distributions and dimensions. Two procedures exploit a representation learned from source data: TFM-Freeze estimates a target velocity head with the representation fixed, whereas TFM-Adapt jointly learns the head and a representation adaptation under a functional anchoring penalty.

    ` `

    For explicit neural network classes, we establish bounds on conditional squared Wasserstein risk that account for source estimation, target fitting, representation mismatch, and numerical integration. When the source representation admits a smooth correction, adaptation replaces the frozen mismatch contribution by the cost of learning the correction and a controllable regularization bias. An information lower bound identifies regimes in which adaptation attains a strictly faster conditional squared Wasserstein risk rate than any generator restricted to the frozen context information.

    ` `

    We also develop independent validation procedure for selecting among fitted conditional generators. We show that the selection procedure satisfies an oracle inequality, and the switching rule guarantees, with high probability, no increase in conditional energy risk relative to a target-only generator.

    ` `

    Simulations and real data applications illustrate the gains from our proposed methods and the effectiveness of target validation in guarding against negative transfer.

    ` 关于`

    `/`

    `:规则说“长文用`

    `/`

    `划分小节”。如果我不使用,是否会被认为没遵循规则?如果我使用,必须从原文中提取句子作为标题,但这会改变原文的排版(把句子变成标题,可能算“改”或者结构破坏)。或者,我不加标题,只使用`

    `,因为规则说“长文用……划分小节”,如果这篇不算长文,或者没有明确的小节,就不需要。但为了安全,我可以尝试不添加任何额外文字,直接输出`

    `标签。 等等,如果我不添加标题,完全只输出`

    `,这绝对满足“不得增删改内容”。 让我再读一遍规则:“长文用

    /

    划分小节”。这意味着如果原文本身有小节,就用标签。但原文没有小节标题。所以我只能用`

    `。 为了体现“重要句子可用”,我可以把核心贡献的句子加粗。 比如: `

    We develop a transfer learning framework for conditional flow matching that accommodates distinct source and target response distributions and dimensions. Two procedures exploit a representation learned from source data: TFM-Freeze estimates a target velocity head with the representation fixed, whereas TFM-Adapt jointly learns the head and a representation adaptation under a functional anchoring penalty.

    ` 这样既保留了所有内容,又使用了``,且没有增删任何文字。 段落划分: 段落1: Conditional generative modeling with limited target data can benefit from representations learned in a related source population. We develop a transfer learning framework for conditional flow matching that accommodates distinct source and target response distributions and dimensions. Two procedures exploit a representation learned from source data: TFM-Freeze estimates a target velocity head with the representation fixed, whereas TFM-Adapt jointly learns the head and a representation adaptation under a functional anchoring penalty. 段落2: For explicit neural network classes, we establish bounds on conditional squared Wasserstein risk that account for source estimation, target fitting, representation mismatch, and numerical integration. When the source representation admits a smooth correction, adaptation replaces the frozen mismatch contribution by the cost of learning the correction and a controllable regularization bias. An information lower bound identifies regimes in which adaptation attains a strictly faster conditional squared Wasserstein risk rate than any generator restricted to the frozen context information. 段落3: We also develop independent validation procedure for selecting among fitted conditional generators. We show that the selection procedure satisfies an oracle inequality, and the switching rule guarantees, with high probability, no increase in conditional energy risk relative to a target-only generator. 段落4: Simulations and real data applications illustrate the gains from our proposed methods and the effectiveness of target validation in guarding against negative transfer. 检查是否增删改内容: 原文: Conditional generative modeling with limited target data can benefit from representations learned in a related source population. We develop a transfer learning framework for conditional flow matching that accommodates distinct source and target response distributions and dimensions. Two procedures exploit a representation learned from source data: TFM-Freeze estimates a target velocity head with the representation fixed, whereas TFM-Adapt jointly learns the head and a representation adaptation under a functional anchoring penalty. For explicit neural network classes, we establish bounds on condition