SIA: Self Improving AI with Harness & Weight Updates
- Published
- Source
- arXiv
- Paper number
- 268
- Field
- AI / General
- arXiv ID
- 2605.27276
Key points
- In SIA-H, which uses only the harness, the agent built a classification pipeline and reached 50.0% accuracy, but performance plateaued once the prompt and code hit their limits.
- In SIA-W+H, which adds weight updates, the model learned to internalize legal nuance, and accuracy jumped to 70.1%, far above the previous best of 45.0%.
- In SIA-W+H, the system achieved a 14.02x speedup over the initial baseline. Here, weight updates were decisive because the model learned specific H100 hardware patterns such as shared-memory tiling and register accumulation, which scaffold iteration alone could not discover.
Paper links
External research summaries. These are not HDATF publications or measured product results.