SenseNova-RoboRSI 用智能体系统设计将 RoboDojo 成绩从 28.97 提升至 56.83
SenseNova 团队不改模型权重，只改智能体系统设计，就让机器人 RoboDojo 分数翻倍，还带递归自我改进循环，做法值得细看。
SenseNova 团队发布 SenseNova-RoboRSI 研究，在不改动 GPT-6 Astra 模型权重的前提下提升具身智能表现。RoboDojo 平均分从基线 28.97 提高到 56.83，任务成功率达 50.83%。该方法在 LIBERO-PRO 上取得 94.50%、在 RoboCasa 上取得 64.60 的成绩。核心技术包括多点末端执行器预测，即一步生成短序列动作目标以提升运动连续性并减少模型调用次数。结合规划与反馈子智能体，主智能体可分解任务、执行动作并根据反馈调整，执行结果还会回灌用于改进未来的智能体配置，形成递归自我改进循环。
A stronger robot doesn't always require a stronger foundation model. Sometimes, it just needs a better way to act.
Recent research from SenseNova team offers an interesting example of how agent system design can improve embodied AI performance.
With GPT-6 Astra's model weights unchanged, SenseNova-RoboRSI reports a RoboDojo average score of 56.83, compared with a published baseline of 28.97, alongside a 50.83% task success rate.
It also reports strong results on LIBERO-PRO (94.50%) and RoboCasa (64.60).
One strategy explored by SenseNova-RoboRSI is multi-point end-effector (EEF) prediction. Rather than predicting and executing one target at a time, it generates short sequences of action targets in a single step, helping improve motion continuity while reducing repeated model calls.
Combined with Planning and Feedback Subagents, the Main Agent can break down tasks, execute actions, monitor progress, and adjust subsequent actions based on feedback.
More importantly, SenseNova-RoboRSI brings these strategies into a recursive self-improvement (RSI) loop, where execution results help guide improvements to future agent harness configurations.
Improving the harness may become just as important as improving the model itself.
- lmarena.ai17:15原文