Tsinghua and NUS Open-Source VISTA: Robots Self-Edit Code, Lifting Success Rate from 17.5% to 95%

1 reports archived1 independent sourcesupdated 11/7/2026, 12:30:00

Synthesis & Latest Updates

1 Sources Cross-Validated

A Tsinghua and National University of Singapore team proposed and open-sourced VISTA, a framework in which a Meta-agent iteratively improves an embodied agent's harness—skill code, invocation logic and persistent memory—through active exploration and code development, without updating model weights. On LIBERO-Pro it raises overall task success from 17.5% to 95.0%, with improvements retained and revisitable across rounds via a harness graph.

LATEST/A Tsinghua and National University of Singapore team proposed and open-sourced VISTA, a framework in which a Meta-agent iteratively improves an embodied agent's harness—skill code, invocation logic and persistent memory—through active exploration and code development, without updating model weights. On LIBERO-Pro it raises overall task success from 17.5% to 95.0%, with improvements retained and revisitable across rounds via a harness graph.

Hourly heat curve

4 fully observed hours

Now

7.7

Peak

8.404:00

24h change

–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first

量子位 (微信公众号)T2Nov 7, 12:30·78 pts
Tsinghua and NUS Open-Source VISTA: Robots Self-Edit Code, Lifting Success Rate from 17.5% to 95%
- VISTA treats the harness—skill code, invocation logic and persistent memory—as the evolution target while model weights stay frozen
- The Meta-agent inspects execution logs, enters simulation to gather extra evidence when needed, and self-tests candidate versions
- Candidates are frozen and independently evaluated, replacing the current version only if better; versions and evidence are stored in a harness graph for rollback
- Overall success on LIBERO-Pro rises from 17.5% to 95.0%
View Related Reports & Analysis Original Source