SenseNova-RoboRSI 以任务反馈改进机器人智能体配置，报告基准成绩

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.

🤖 𝗘𝘃𝗲𝗿𝘆 𝗿𝗼𝗯𝗼𝘁 𝘁𝗮𝘀𝗸 𝗳𝗮𝗶𝗹𝘂𝗿𝗲 𝗰𝗮𝗻 𝗯𝗲𝗰𝗼𝗺𝗲 𝗮 𝗹𝗲𝘀𝘀𝗼𝗻 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗻𝗲𝘅𝘁 𝗶𝘁𝗲𝗿𝗮𝘁𝗶𝗼𝗻. 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗥𝗼𝗯𝗼𝗥𝗦𝗜 uses 𝗥𝗦𝗜 to evolve 𝗮𝗴𝗲𝗻𝘁 𝗵𝗮𝗿𝗻𝗲𝘀𝘀𝗲𝘀 through physical task feedback, without updating model parameters. ▶ 𝗥𝗼𝗯𝗼𝗗𝗼𝗷𝗼: 56.83 avg. score, 50.83% success rate, +𝟮𝟳.𝟴𝟲 𝗽𝗼𝗶𝗻𝘁𝘀 𝘃𝘀. 𝘁𝗵𝗲 𝗚𝗣𝗧-𝟲 𝗔𝘀𝘁𝗿𝗮 𝗯𝗮𝘀𝗲𝗹𝗶𝗻𝗲 (𝟮𝟴.𝟵𝟳) ▶ 𝗥𝗦𝗜 𝗹𝗼𝗼𝗽: Execute → Diagnose → Improve → Validate → Inherit ▶ 𝗘𝘅𝗽𝗹𝗼𝗿𝗲𝗱 𝗮𝗻𝗱 𝘃𝗮𝗹𝗶𝗱𝗮𝘁𝗲𝗱: Multi-point EEF prediction and Planning + Feedback Subagents This achievement reflects the collective efforts of our joint team. RoboRSI will be 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲𝗱 𝘀𝗼𝗼𝗻 at https://t.co/iTVQ1YDFcn. A comprehensive 𝘁𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗿𝗲𝗽𝗼𝗿𝘁 detailing our methods and findings is also on the way. Stay tuned!