外观
论文arXiv 2610.0610423 天前
条件轨迹峰：动作块上的单次多模态策略
原文链接：Conditional Trajectory Peaks: Single-Pass Multimodal Policies over Action Chunks（arXiv 2610.06104 · 发布日期：2026-10-05）
Conditional Trajectory Peaks: Single-Pass Multimodal Policies over Action Chunks
CTP面向多模态模仿学习，用单次前向预测动作块候选、概率质量与尺度，并在多个基准和真实双臂任务上验证。
核心要点
- 多模态模仿学习需在同一观测下给出多样可执行未来，并在重规划间保持一致行为。
- CTP联合预测完整动作块候选、概率质量和轨迹尺度，以单次策略框架建模多模态轨迹。
- DAPS用轨迹级后验责任和重叠约束专化轨迹峰，ETBT通过几何对应保持跨块一致性。
- CTP在Push-T、D3IL、LIBERO取得较高成功率；真实双臂任务中保留双放置模式并降低推理延迟。
验证范围与局限：验证范围限于摘要列出的 Push-T、D3IL、LIBERO 基准及真实双臂两盘、扶瓶和放笔任务。
英文摘要 (Abstract)
Multimodal imitation learning requires diverse executable futures under the same observation and consistent behavior across replanning cycles. We present Conditional Trajectory Peaks (CTP), a single-pass policy framework that jointly predicts complete action-chunk candidates, probability masses, and trajectory scales. Distribution-Aware Peak Specialization (DAPS) specializes trajectory peaks using trajectory-level posterior responsibilities and mass- and scale-modulated overlap constraints. Evidence-Gated Trajectory Belief Transport (ETBT) maintains cross-chunk consistency through geometric correspondence between exchangeable candidate sets, while allowing current policy evidence to override historical constraints. CTP achieves a coverage score of 91.40% on Push-T; success rates of 100.0%, 79.72%, and 84.44% on D3IL Avoiding, Aligning, and Sorting-2, respectively. On LIBERO, CTP achieves an average success rate of 97.25%. In real-world dual-arm experiments, CTP preserves both placement modes in a two-plate task, succeeding in all 50 trials. On bottle uprighting and pen placement into a holder, it maintains success rates comparable to $π_{0.5}$ while reducing policy inference latency from 218.24 ms to 75.80 ms. These results demonstrate that single-pass trajectory modeling can combine multimodal behavior, closed-loop consistency, and efficient inference.
作者：Di Wu, Rongtian Shen, Ping Liu, Xuhua Chen, He Zheng, Lingfeng Zhang, Tao Zhang
来源
本页事实对应的公开页面，以原文为准。