VISTA Framework Enables Self-Improving Robots Without Updating Model Weights
Overview
Researchers from Tsinghua University and the National University of Singapore have introduced VISTA, a framework that allows robots to autonomously develop and refine their own skills through code modification. By employing a Meta-agent to iteratively improve Harnesses while keeping underlying model parameters fixed, the system increased task success rates on the LIBERO-Pro benchmark from 17.5% to 95.0%. The approach has been validated in simulation environments, with source code and papers available for public use.
Generated from attributed reports · Updated 18 hours ago
Event evidence and corrections
1 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · Reported success rate: 17.5 percent · Basis not reported · Differing source assertions
“在底层模型参数保持固定的情况下，VISTA将LIBERO-Pro的整体任务成功率从 17.5% 提升到了 95.0% 。”
Exact source · revision 1qbitai
Reported quantity · Reported success rate: 95 percent · Basis not reported · Differing source assertions
“在底层模型参数保持固定的情况下，VISTA将LIBERO-Pro的整体任务成功率从 17.5% 提升到了 95.0% 。”
Exact source · revision 1qbitai
Reported quantity · Reported success rate: 22.8 percent · Basis not reported · Differing source assertions
“LIBERO-T和LIBERO-S的平均成功率分别从22.8%、12.2%提升至96.6%、93.4%；RoboTwin的50个任务从48.5%提升至74.0%。”
Exact source · revision 1qbitai
Reported quantity · Reported success rate: 12.2 percent · Basis not reported · Differing source assertions
“LIBERO-T和LIBERO-S的平均成功率分别从22.8%、12.2%提升至96.6%、93.4%；RoboTwin的50个任务从48.5%提升至74.0%。”
Exact source · revision 1qbitai
Reported quantity · Reported success rate: 96.6 percent · Basis not reported · Differing source assertions
“LIBERO-T和LIBERO-S的平均成功率分别从22.8%、12.2%提升至96.6%、93.4%；RoboTwin的50个任务从48.5%提升至74.0%。”
Exact source · revision 1qbitai
Reported quantity · Reported success rate: 93.4 percent · Basis not reported · Differing source assertions
“LIBERO-T和LIBERO-S的平均成功率分别从22.8%、12.2%提升至96.6%、93.4%；RoboTwin的50个任务从48.5%提升至74.0%。”
Exact source · revision 1qbitai
Report timeline
Follow attributed reports and material updates.
- QbitAI — Robotics on WeChatSignalVISTA Framework Enables Self-Improving Robots Without Updating Model Weights
Researchers from Tsinghua University and the National University of Singapore have introduced VISTA, a framework that allows robots to autonomously develop and refine their own skills through code modification. By employing a Meta-agent to iteratively improve Harnesses while keeping underlying model parameters fixed, the system increased task success rates on the LIBERO-Pro benchmark from 17.5% to 95.0%. The approach has been validated in simulation environments, with source code and papers available for public use.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).