让视觉语言动作模型按需计算，复杂任务成功率超90%。
Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoning
- 用隐空间迭代优化替代显式推理，实现动态计算深度。
- 四次迭代下复杂任务成功率突破90%，简单任务快速收敛。
- 内存恒定、推理快80倍，适合机器人实时决策场景。
当前视觉-语言-动作（VLA）模型采用固定计算深度，对简单调整与复杂多步操作耗用相同算力。虽然思维链（CoT）提示可变计算量，但内存线性增长且不适用于连续动作空间。本文提出递归深度VLA（RD-VLA），通过隐空间迭代精炼实现计算自适应，而非显式生成文本。该模型采用循环权重共享的动作头，支持任意推理深度且内存恒定。训练使用截断时间反向传播（TBPTT）监督精炼过程。推理时，基于隐状态收敛性动态决定停止时机。在挑战性操作任务上验证：单次推理失败（0%成功率）的任务，四次迭代后成功率超过90%；简单任务迅速饱和。RD-VLA为机器人测试时计算扩展提供可扩展路径，以隐空间推理替代文本推理，实现恒定内存和最高达80倍的推理速度提升。
原文摘要 · Abstract (English)
Current Vision-Language-Action (VLA) models rely on fixed computational depth, expending the same amount of compute on simple adjustments and complex multi-step manipulation. While Chain-of-Thought (CoT) prompting enables variable computation, it scales memory linearly and is ill-suited for continuous action spaces. We introduce Recurrent-Depth VLA (RD-VLA), an architecture that achieves computational adaptivity via latent iterative refinement rather than explicit token generation. RD-VLA employs a recurrent, weight-tied action head that supports arbitrary inference depth with a constant memory footprint. The model is trained using truncated backpropagation through time (TBPTT) to efficiently supervise the refinement process. At inference, RD-VLA dynamically allocates compute using an adaptive stopping criterion based on latent convergence. Experiments on challenging manipulation tasks show that recurrent depth is critical: tasks that fail entirely (0 percent success) with single-iteration inference exceed 90 percent success with four iterations, while simpler tasks saturate rapidly. RD-VLA provides a scalable path to test-time compute in robotics, replacing token-based reasoning with latent reasoning to achieve constant memory usage and up to 80x inference speedup over prior reasoning-based VLA models. Project page: https://rd-vla.github.io/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。