Hyperagents

2026-03-19

AI Summary

Current self-improving AI systems are constrained by fixed, handcrafted meta-mechanisms, limiting their capacity for open-ended self-enhancement beyond code-centric domains. This work proposes the Hyperagents framework, which unifies task agents and editable meta-agents into a single modifiable program, thereby enabling, for the first time, the editability of metacognitive mechanisms themselves. This breakthrough removes the requirement that task performance and self-improvement capabilities be aligned within the same domain. Built upon an extension of the Darwinian Gödel Machine (DGM), the resulting DGM-Hyperagents (DGM-H) integrates editable metacognition, persistent memory, and performance tracking, achieving sustained acceleration in self-improvement across diverse tasks. Empirical results demonstrate significant outperformance over existing baselines and reveal emergent cross-domain transfer and cumulative gains at the meta-level.

Technology Category

• Search and Optimization: Metareasoning and Metaheuristics
• Multiagent Systems: Adversarial Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Self-improving AI systems aim to reduce reliance on human engineering by learning to improve their own learning and problem-solving processes. Existing approaches to self-improvement rely on fixed, handcrafted meta-level mechanisms, fundamentally limiting how fast such systems can improve. The Darwin Gödel Machine (DGM) demonstrates open-ended self-improvement in coding by repeatedly generating and evaluating self-modified variants. Because both evaluation and self-modification are coding tasks, gains in coding ability can translate into gains in self-improvement ability. However, this alignment does not generally hold beyond coding domains. We introduce \textbf{hyperagents}, self-referential agents that integrate a task agent (which solves the target task) and a meta agent (which modifies itself and the task agent) into a single editable program. Crucially, the meta-level modification procedure is itself editable, enabling metacognitive self-modification, improving not only the task-solving behavior, but also the mechanism that generates future improvements. We instantiate this framework by extending DGM to create DGM-Hyperagents (DGM-H), eliminating the assumption of domain-specific alignment between task performance and self-modification skill to potentially support self-accelerating progress on any computable task. Across diverse domains, the DGM-H improves performance over time and outperforms baselines without self-improvement or open-ended exploration, as well as prior self-improving systems. Furthermore, the DGM-H improves the process by which it generates new agents (e.g., persistent memory, performance tracking), and these meta-level improvements transfer across domains and accumulate across runs. DGM-Hyperagents offer a glimpse of open-ended AI systems that do not merely search for better solutions, but continually improve their search for how to improve.

Problem

• self-improving AI
• open-ended improvement
• meta-level modification
• hyperagents
• computable tasks

Innovation

• hyperagents
• self-improving AI
• metacognitive self-modification
• open-ended learning
• editable meta-mechanism