AI 中文总结
提出无需训练的测试时控制框架DART-VLN，通过记忆衰减和反循环正则化改善离散视觉语言导航的局部回溯和历史证据过时问题，提升路径效率与导航性能。
AI 中文摘要
基于记忆的离散视觉语言导航（VLN）代理必须在部分可观测性下行动，但即使强大的冻结骨干网络在测试时仍然脆弱。两种常见的失败模式是记忆读取时过时的历史证据以及动作选择期间低效的局部回溯。我们提出了DART-VLN，一种针对离散VLN的无需训练的测试时控制框架。DART-VLN结合了测试时记忆衰减（一种读取侧记忆重加权规则，在不重写存储内容的情况下抑制过时和冗余的证据）与反循环正则化（一种轻量级的下一跳惩罚，在动作选择期间阻止立即反转）。该框架不引入新的可学习参数，并保持学习到的骨干网络不变。在R2R和REVERIE上的实验显示出一致的模式：仅衰减提供稳定的读取侧增益，而衰减+反循环实现了最佳的整体质量-效率权衡，在关键设置下产生更短的轨迹、更低的运行时间和改进的导航性能。行为分析进一步证实，反循环正则化减少了局部回溯，并在冻结骨干网络下提高了路径效率。总体而言，结果表明适度的测试时控制可以使基于记忆的离散VLN更可靠和高效，而无需重新训练。
英文摘要
Memory-based agents for discrete vision-language navigation (VLN) operate under partial observability and may exhibit systematic inference-time failures even with strong pretrained backbones. This paper addresses two recurring problems: stale historical evidence during memory readout and inefficient local backtracking during action selection. We present DART-VLN, a training-free inference-time framework that combines Test-Time Memory Decay, which reweights stale and redundant memory slots without modifying their stored content, with Anti-Loop Regularization, a lightweight next-hop penalty that discourages immediate reversals. DART-VLN introduces no learnable parameters and leaves the navigation backbone unchanged. Experiments on R2R and REVERIE showed that memory decay preserved or improved task performance while reducing runtime. The addition of anti-loop regularization further shortened trajectories, reduced local backtracking, and achieved the best overall balance between navigation quality and efficiency among the evaluated GridMM variants. These results indicate that lightweight inference-time control can improve the reliability and efficiency of memory-based discrete VLN without retraining.
CommentsAccepted by the 2026 IEEE International Conference on Systems, Man, and Cybernetics (IEEE SMC 2026)
