10 pages tagged

#self-improvement

agentic-context-engineering concept ACE: context as evolving playbook; Generator/Reflector/Curator; incremental bullets only. context-engineering concept Choosing what enters the LLM context; steering-not-teaching; tiers, Context Rot, compaction. evolutionary-search-llm concept Evolutionary optimization for LLM programs: AlphaEvolve, DGM, Promptbreeder, ADAS, AFlow. harness concept Acceptance + boundary + signal + fallback around the loop; matters more than model choice. meta-context-engineering concept MCE: bi-level search over context *mechanism* + content; free-form skills, agentic crossover. meta-harness concept Outer-loop optimizer that rewrites harness *programs*; same-model 6× spread. recursive-self-improvement concept RSI: AI improving its own machinery; the near-term path is harness optimization, not weights. reward-hacking concept Reward overfitting → hacking → tampering → alignment faking; acute in self-improvement loops. self-improving-harness concept An LLM optimizing its own harness: STOP, Self-Harness, DGM; propose-evaluate-accept loops. self-reflection concept Reflexion / Chain of Hindsight / Algorithm Distillation; ancestor of verifier + ralph-wiggum loops.
esc