ADSD Framework Uses Auto-Diagnosis to Cut Numerical Solver Error by 71x
Peter Chen · hf · 2026-10-06
ADSD (Auto-Diagnosis and Skill Discovery) tackles the gap between generating scientific code and improving the algorithms behind it.
- Diagnosis-first paradigm: Execution feedback exposes poor solver performance but rarely its cause; ADSD first explains why a solver underperforms, then uses that diagnosis to discover appropriate numerical methods.
- Reusable skills: Discovered knowledge is packaged into reusable solver skills, turning improvement from trial-and-error editing into a structured diagnosis-discovery-implementation process.
Across power flow equations, AC optimal power flow control, stiff ODEs, and heterogeneous diffusion PDEs, ADSD consistently improves accuracy, robustness and efficiency—reducing mean solver error on GOC-500 power flow by nearly 71x, with gains transferring to unseen grid topologies.
More from Research
- Subsampling and extrapolation keep the Mandelbrot area estimate unbiased near the boundary — geoffreyirving · 2026-10-06
- New estimate sits 6.5e-9 below Hsing Lo's 2025 value; reproduction suggests the gap is a fluctuation — geoffreyirving · 2026-10-06
- Claude-assisted CUDA compute pins Mandelbrot set area to 1.506591883653, 60x tighter than 2012 record — geoffreyirving · 2026-10-06
- Group-Evolving Agents: a new paradigm where the unit of agent self-improvement is a group — xwang_lk · 2026-10-06
- SLIM paper at COLM: design principles for long-horizon agentic search systems — xiye_nlp · 2026-10-06
- The Nobel optogenetics drama: forgotten inventor Zhuo-Hua Pan had the stronger claim — _onionesque · 2026-10-06