AI systems increasingly participate in their own improvement: revising their outputs, adapting their own harnesses during deployment, training on data they generate, and, increasingly, conducting AI research itself. This literature is described under a vocabulary ("self-refine," "self-reward," "self-play," "self-evolve") that conflates fundamentally different ambitions.
arXiv:2607. 07663v1 Announce Type: new Abstract: AI systems increasingly participate in their own improvement: revising their outputs, adapting their own harnesses during deployment, training on data they generate, and, increasingly, conducting AI research itself.
arXiv:2609.06396v2 Announce Type: new Abstract: Recursive self-improvement (RSI) lets a system improve the model-building machinery from its own failures, so every later model inherits the gain. Yet...
The paper introduces AIDE^2, an AI research agent that recursively improves its own code by proposing, benchmarking, and selecting modifications. Over an eight‑day autonomous run, it achieved seven successive improvements—including new search policies and memory mechanisms—that transferred to four held‑out benchmarks in machine learning, algorithm engineering, and weather forecasting. The agent’s best version matched or outperformed a top human‑engineered production research agent and also reduced reward‑hacking rates, despite never optimizing for that metric.
arXiv:2609.00137v1 Announce Type: new Abstract: AI is increasingly used in the R\&D process that produces future AI systems. We study the conditions under which this feedback becomes self-amplifying....
arXiv:2607. 05297v1 Announce Type: new Abstract: Recent LLM agents tackle increasingly long-horizon, open-ended tasks, and external skills, reusable procedural knowledge supplied to the agent, further extend this capability.
The paper introduces Generalized Agent Iteration (GAI), a formal framework that unifies iterative policy improvement and recursive self‑improvement (RSI) under a single learning paradigm. GAI treats an agent as a configuration of modifiable components and models learning as a cycle of evaluation and improvement, with two key dials: whether the improving mechanism is part of the agent and whether the evaluation standard is external. These dials distinguish between generalized policy iteration (GPI) and RSI, and classify systems as anchored, goal‑drift, or fully self‑referential, allowing existing systems to be mapped and RSI defects to be analyzed systematically.
arXiv:2607. 15524v1 Announce Type: cross Abstract: Under model--harness co-evolution, harnesses are not merely inference-time scaffolds but data-generating components whose execution traces can shape future foundation models.
arXiv:2609.35561v2 Announce Type: replace Abstract: Recursive self-improvement (RSI) seeks to enable AI systems to participate in improving their own capabilities. A concrete pathway is autonomous mo...
arXiv:2609.35897v1 Announce Type: new Abstract: The pursuit of recursive self-improvement (RSI) toward general intelligence is divided between macro-level language model scaling and the interaction-d...
arXiv:2606. 20657v2 Announce Type: replace Abstract: Post-training a frontier model is normally weeks of human work: proposing data and recipe changes, launching runs, reading evals, deciding what to keep.
The paper discusses recursive self‑improvement (RSI) for AI, describing how systems can use experience and feedback to make lasting enhancements to both their abilities and their future improvement processes. It introduces the Headroom‑Closed Index (HCI) to expose limitations in current large language models and outlines a development roadmap for RSI, progressing from autonomy in execution to full recursive meta‑improvement. The authors analyze RSI in various contexts such as scientific discovery, embodied intelligence, and software engineering, noting differing requirements and development speeds, and they connect RSI research to practical industry applications while highlighting key challenges to achieving genuine RSI.
The paper introduces iCoder-27B, a 27‑billion‑parameter model for RTL design and GPU kernel optimization that is developed through a recursive AI‑led process with minimal human input. Human experts provide high‑level objectives and reusable research skills, while the agent autonomously selects experiments, diagnoses outcomes, and refines training strategies, coordinating SFT, self‑distillation, and reinforcement learning. iCoder outperforms GPT‑5.5 and Claude‑Opus‑4.8 on several benchmarks, demonstrating the feasibility of building frontier‑competitive models with largely automated development.