Safety
LabEvolver: Training-Free Experience Evolution for Safe and Grounded Wet-Lab Agents
arXiv:2607.27690v1 Announce Type: new Abstract: We introduce LabEvolver, a training-free framework that equips safe and grounded wet-lab agents with episodic memory from execution experience. LabEvolv
arXiv:2607.27690v1 Announce Type: new Abstract: We introduce LabEvolver, a training-free framework that equips safe and grounded wet-lab agents with episodic memory from execution experience. LabEvolver couples a state-grounded inner trial loop for adaptive perception, online planning, and safety validation with an outer evolution loop that distills completed trajectories into reusable skill, strategy, and safety experience. On robotic solution-preparation tasks, LabEvolver demonstrates real-world feasibility, reducing pH-regulation completion time and safety-gate intercepts by 48.2% and 60.0%, respectively. On ALFWorld, it further improves cumulative success rate within 20 steps from 76.2% with ReAct to 91.4% over 500 continual tasks, showing generality beyond wet-lab settings. These results support learn-by-doing experience evolution as a feasible path toward closed-loop automated scientific discovery. The project page is available at https://github.com/AndyGao6186/LabEvolver.
Related
- Long-Term Memory for VLA-based Agents in Open-World Task Execution
- Governed Capability Evolution for Embodied Agents: Safe Upgrade, Compatibility Checking, and Runtime Rollback for Embodied Capability Modules
- Analytic Concept-Centric Memory for Agentic Embodied Manipulation
- Learning Without Losing Identity: Capability Evolution for Embodied Agents
Source: arXiv cs.RO | 2026-07-31