首页 > AI前沿 > WoRD: Diffusion-Based Posterior Inference for Online Goal Recognition

WoRD: Diffusion-Based Posterior Inference for Online Goal Recognition

arXiv机器学习 2026-11-18 02:35 9 阅读 查看原文
用户要求将提供的纯文本新闻正文排版成语义化的HTML片段。 需要遵循以下规则: 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。
    
    提供的正文:
    "Goal recognition is the task of inferring an agent's intended goal from its observed behavior. Most existing work focuses on high-level symbolic recognition, abstracting away the continuous motion through which agents reach their goal. Methods that do reason over continuous trajectories rely on an explicit dynamics model that is hard to acquire and rarely transfers across settings. Neither approach explicitly learns how agents move toward each candidate goal, limiting accuracy on ambiguous trajectory prefixes where such knowledge is most needed. We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations. At each step, WoRD runs one warm-started denoising chain per candidate goal, steers each chain toward the observed trajectory through a closed-form log-likelihood gradient, and reweights the resulting samples by Bayes' rule into a calibrated posterior over goals. We evaluate WoRD on simulated robot settings, pedestrian trajectories, and vehicle trajectories using four metrics targeting distinct failure modes. Across all three settings, the same inference procedure preserves calibrated multimodal beliefs through ambiguous prefixes where filter- and classifier-based observers collapse to a single mode or grow overconfident."
    
    分析正文内容,它实际上是一段学术摘要,没有明显的列表、引用或代码。可以将其分为几个小节,或者就一个段落。但规则提到“长文用

    /

    划分小节”,这里虽然不算特别长,但为了体现结构,可以划分小节,比如“背景与现有方法”、“WoRD方法”、“实验与评估”。不过规则4要求“不得增删改内容”,如果我加小标题,算不算“增删改内容”?“保留原文所有信息,不得增删改内容”通常指文本内容本身,但加上HTML标签是允许的。不过如果加小标题的文字,就是增加了内容。为了安全起见,不添加原文没有的小标题文字,只使用`

    `或`

    `包裹已有的句子,或者将段落用`

    `包裹,如果不需要划分小节就不强行划分,或者把第一句作为小节标题? 原文本只有一段。如果我把它拆成几个`

    `,只要不改变文字即可。 比如: `

    Goal recognition is the task of inferring an agent's intended goal from its observed behavior. Most existing work focuses on high-level symbolic recognition, abstracting away the continuous motion through which agents reach their goal. Methods that do reason over continuous trajectories rely on an explicit dynamics model that is hard to acquire and rarely transfers across settings. Neither approach explicitly learns how agents move toward each candidate goal, limiting accuracy on ambiguous trajectory prefixes where such knowledge is most needed.

    ` `

    We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations. At each step, WoRD runs one warm-started denoising chain per candidate goal, steers each chain toward the observed trajectory through a closed-form log-likelihood gradient, and reweights the resulting samples by Bayes' rule into a calibrated posterior over goals.

    ` `

    We evaluate WoRD on simulated robot settings, pedestrian trajectories, and vehicle trajectories using four metrics targeting distinct failure modes. Across all three settings, the same inference procedure preserves calibrated multimodal beliefs through ambiguous prefixes where filter- and classifier-based observers collapse to a single mode or grow overconfident.

    ` 这样既符合“每个自然段落用

    包裹”,又没有增加任何额外文字。原文本其实是一个长段落,我可以根据语义把它分成几个自然段,这不违反“不得增删改内容”(内容指文字本身)。或者保留为一个`

    `。分成几个`

    `更好看,也符合排版要求。 检查是否需要`

    `/`

    `:规则说“长文用

    /

    划分小节”,这段不算长文(只有约200字),所以可以不用小标题,或者用`

    `把核心方法提出来?不,不加文字的话没法用标题。如果强行把第一句放`

    `,语义可能不对。所以直接用`

    `包裹即可。 重要句子可用``。比如把提出WoRD的句子,或者最后一句结论加粗。 "We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations." 可以加粗。 排版如下:

    Goal recognition is the task of inferring an agent's intended goal from its observed behavior. Most existing work focuses on high-level symbolic recognition, abstracting away the continuous motion through which agents reach their goal. Methods that do reason over continuous trajectories rely on an explicit dynamics model that is hard to acquire and rarely transfers across settings. Neither approach explicitly learns how agents move toward each candidate goal, limiting accuracy on ambiguous trajectory prefixes where such knowledge is most needed.

    We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations. At each step, WoRD runs one warm-started denoising chain per candidate goal, steers each chain toward the observed trajectory through a closed-form log-likelihood gradient, and reweights the resulting samples by Bayes' rule into a calibrated posterior over goals.

    We evaluate WoRD on simulated robot settings, pedestrian trajectories, and vehicle trajectories using four metrics targeting distinct failure modes. Across all three settings, the same inference procedure preserves calibrated multimodal beliefs through ambiguous prefixes where filter- and classifier-based observers collapse to a single mode or grow overconfident.

    检查是否满足所有规则: 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(分为三个段落,非长文不需要h标签,如果非要用,可以把“背景”、“方法”、“评估”加h3,但会增字,所以不用。) 3. 列表、引用、代码按需,重要句子用。 -> 满足,用了一个。 4. 保留原文所有信息,不得增删改内容。 -> 满足,文字一字未改,只是分了段和加了标签。 5. 不要输出文档级标签。 -> 满足。 再审视一下分段落是否算“增删改内容”。通常排版允许合理分段。原文本是一个连续的段落,我将其按语义分成三个段落,文字内容没有任何增删改。如果严格保留原样,也可以只输出一个`

    `。为了稳妥,按照语义分段是排版工程师的常规操作。 最终输出:

    Goal recognition is the task of inferring an agent's intended goal from its observed behavior. Most existing work focuses on high-level symbolic recognition, abstracting away the continuous motion through which agents reach their goal. Methods that do reason over continuous trajectories rely on an explicit dynamics model that is hard to acquire and rarely transfers across settings. Neither approach explicitly learns how agents move toward each candidate goal, limiting accuracy on ambiguous trajectory prefixes where such knowledge is most needed.

    We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations. At each step, WoRD runs one warm-started denoising chain per candidate goal, steers each chain toward the observed trajectory through a closed-form log-likelihood gradient, and reweights the resulting samples by Bayes' rule into a calibrated posterior over goals.

    We evaluate WoRD on simulated robot settings, pedestrian trajectories, and vehicle trajectories using four metrics targeting distinct failure modes. Across all three settings, the same inference procedure preserves calib