Knowledge-Centric Self-Improvement

2026-07-21

AI Summary

This work addresses the limitations of traditional agent self-improvement methods, which are often constrained by task-specific designs or narrow task distributions, leading to poor transferability and high maintenance costs. The authors propose a knowledge-centric self-improvement paradigm that shifts the focus from the agent itself to a structured, transferable, and auditable knowledge base. In this framework, a general-purpose agent contributes insights to a shared knowledge repository after task completion through an evidence-driven mechanism, enabling continuous refinement via intra- and cross-task forums, knowledge distillation, and collaborative reasoning with large language models. Empirical results demonstrate substantial improvements in solution success rates and reduced computational overhead across abstract reasoning, programming, and terminal tasks. Moreover, the distilled knowledge effectively generalizes to unseen tasks and diverse families of large language models.

Abstract

Self-improving AI systems typically treat the agent as the object that improves, by optimizing prompts, workflows, harnesses, or even the agent's own code. This agent-centric view can make improvements expensive to maintain and difficult to transfer, because gains become tied to a particular agent design, task distribution, or adaptation run. We study a complementary paradigm: knowledge-centric self-improvement, in which agents remain generic and disposable while the persistent object is a curated knowledge base that agents can leverage for future tasks. We conduct controlled case studies to operationalize this idea via a simple protocol. Agents attempt one task, then contribute evidence-grounded insights to a shared knowledge base via task-level and cross-task forums, followed by knowledge distillation. Because self-improvement is contained in the knowledge rather than the agent, improvement can be more inspectable, transferable, and portable. Across abstract reasoning, coding, and terminal benchmarks, this protocol improves solve rates while reducing dollar cost relative to agent-centric baselines. The resulting distilled knowledge also transfers to held-out tasks and across LLM families, indicating that the improvement is not merely an LLM- or run-specific behavior. These results support a new view of self-improving agentic systems: progress can be driven primarily by the curated persistent knowledge. Code is available at https://github.com/recursive-knowledge/KSI.

Problem

• self-improving AI
• agent-centric improvement
• knowledge-centric paradigm
• transferability
• maintainability

Innovation

• knowledge-centric
• self-improvement
• knowledge distillation
• transferable knowledge
• agentic systems