The Key Insight: Dosage Depends on Capability Learning happens around the model, not inside it Results Across the Spectrum The three configurations we compare The three patterns, in one view The Cheapest Memory Strategy Can Also Be the Best Memory Should Be Calibrated, Not Merely Accumulated What's Next Appendix: Understanding the Metrics In our previous post, we compared ALTK-Evolve with ACE and showed that how you deliver an agent's self-distilled guidelines — a few retrieved per task vs. the whole set injected — drives both accuracy and cost. This post steps back to the question that comes before it: how much should you give it?
Agentic memory is not a feature you switch on. It's a dose you calibrate to the model.
