Mendel Gödel Machine: Recursive Self-Improving Agents via Comparative Evolution
burny_tech · x · 2026-08-14
The paper introduces Mendel Gödel Machine (MGM), a method for recursive self-improving coding agents via comparative evolution.
- Problem: Self-improving agents typically edit based on a single failure.
- Method: MGM introduces reaction-norm mutation and cross-lineage hybridization to leverage comparative signals.
- Results: This approach yields a +10% improvement on SWE-bench and +42% on Polyglot.
Related event: Mendel Gödel Machine: Self-Improving Agents via Comparative Evolution(2 posts)→
More from coding & agent
- Devs say LLM coding mistakes are a different class: overcomplication, inferred limits, worse at higher reasoning — Mu5_ · 2026-10-16
- How do you debug bad agent runs when you already have tracing? — Sensitive-Parsnip-12 · 2026-10-16
- Where does pre-task context live for agents? Devs debate workflows beyond handoff — shishen2026 · 2026-10-16
- No-instruction autonomous agent wakes up, sets own goals, and tries hacking by default — sethlazar · 2026-10-16
- CODESTRUCT reframes codebases as AST action spaces to fix brittle LLM code agents — burkov · 2026-10-16
- Would you install another SDK for agent debugging on top of your tracing stack? — Sensitive-Parsnip-12 · 2026-10-16