NEWS · RESEARCH · #129
GeoSkill: Experience-Driven Hierarchical Skill Learning with Collaborative Revision for Geospatial Agents
This arXiv paper proposes GeoSkill, a framework that distills historical execution experience of geospatial agents into a Hierarchical Skill Bank (separate Planning and Tool Skill Banks) and applies a Collaborative Trace-driven Skill Revision (CTSR) process — involving Judge, Critic, and Refiner roles — to identify and fix skill defects. The authors report that GeoSkill learns and validates skills during development, freezes the skill bank for retrieval-only deployment, and improves end-to-end task accuracy and tool-execution reliability on EarthBench and ThinkGeo.
KEY POINTS
- This arXiv paper proposes GeoSkill, a framework that distills historical execution experience of geospatial agents into a Hierarchical Skill Bank (separate Planning and Tool Skill Banks) and applies a Collaborative Trace-driven Skill Revision (CTSR) process — involving Judge, Critic, and Refiner roles — to identify and fix skill defects.
- The authors report that GeoSkill learns and validates skills during development, freezes the skill bank for retrieval-only deployment, and improves end-to-end task accuracy and tool-execution reliability on EarthBench and ThinkGeo.
- GeoSkill targets a practical limitation in LLM-augmented geospatial agents — summarizing long-horizon tool chains and avoiding misattributed self-revisions — which can make agent behavior more reusable and reliable across tasks.
WHY IT MATTERS
GeoSkill targets a practical limitation in LLM-augmented geospatial agents — summarizing long-horizon tool chains and avoiding misattributed self-revisions — which can make agent behavior more reusable and reliable across tasks.