SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent
- Published
- Source
- arXiv
- Paper number
- 852
- Field
- AI / General
- arXiv ID
- 2608.07449
Key points
- It formulates skill modification as a forward-backward optimization problem. The forward pass verifies diagnosis, revision, and execution, while the backward pass estimates the usefulness of each knowledge unit for integration or removal.
- It introduces a result-based verification loop that reruns the same task after revision and rolls back if performance drops.
- In the backward pass, it measures the contribution of each knowledge unit with a leave-one-out scheme, and removing bad knowledge improves accuracy from 46 percent to 54 percent.
- Across three backbone LLMs, it achieves an average accuracy gain of 3.0 percentage points over the previous best method.
- Unlike prior methods that only keep adding skills, its key distinction is that it explicitly asks what should be removed.
Paper links
External research summaries. These are not HDATF publications or measured product results.