research
The Skill Graph Tested Whether Its Edges Mattered
CaSKG uses counterfactual probes before publishing a graph for compact procedural retrieval.
Summary
CaSKG uses counterfactual probes before publishing a graph for compact procedural retrieval.
Across six model backbones and two embodied-agent benchmarks, CaSKG led all twelve model-benchmark combinations. Against Graph-of-Skills, the reported macro-average rose from 72.62 to 80.50 on ScienceWorld and from 80.01 to 86.79 percent success on ALFWorld, while mean environment steps also fell.
Why it matters
CaSKG uses counterfactual probes before publishing a graph for compact procedural retrieval.
Limits and context
No additional limitation was separately recorded.
Key claims
CaSKG uses counterfactual probes before publishing a graph for compact procedural retrieval.
Evidence: source-2026-08-27-011
Sources
- arXiv preprint 2608.25500arXiv · primary research
Corrections
No corrections have been recorded for this story.