The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement
cs.LG, cs.AI, cs.CL
Submitted: 2026-09-10
Updated: 2026-09-22
Code: https://github.com/theseus-labs-rsi/awesome-rsi
Project page: https://theseus-labs-rsi.github.io
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
The gist: Recursive self-improvement (RSI) enables AI systems to turn experience and feedback into persistent changes that improve both their capabilities and the process of future improvement.
Terminology
Abstract
Recursive self-improvement (RSI) enables AI systems to turn experience and feedback into persistent changes that improve both their capabilities and the process of future improvement. We first use the Headroom-Closed Index (HCI) to reveal the problems of existing LLMs, then introduce the RSI concept and its development roadmap: from improvement-execution autonomy, improvement-strategy autonomy, experience-acquisition autonomy, and environment-adaptation autonomy, to recursive meta-improvement. Next we examine RSI across scenarios (e.g., scientific discovery, embodied intelligence, software engineering), highlighting their distinct requirements and development speeds. Drawing on diverse industry practices and preliminary empirical evidence, we connect RSI research with practical systems and identify key challenges to achieving genuine RSI.
Sources
- AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning dataset
- Humanity's Last Exam
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- Agentic Neural Architecture Search
- DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
- Goedel Machines: Self-Referential Universal Problem Solvers Making Provably Optimal Self-Improvements
- Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation
- A-Evolve-Training: Autonomous Post-Training of a 30B Model
- Ouroboros: A Self-Developing Frontier Coding Agent with Reviewed Core Evolution
- The Red Queen G\"odel Machine: Co-Evolving Agents and Their Evaluators
- Self-Harness: Harnesses That Improve Themselves
- SIMA 2: A Generalist Embodied Agent for Virtual Worlds
- PANDO: Efficient Multimodal AI Agents via Online Skill Distillation
- A Survey on Self-Evolution of Large Language Models
- A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems
- Self-Improvements in Modern Agentic Systems: A Survey
- Co-Evolution in Agentic Systems: Toward Self-Directed Evolution Beyond Human Design
- A Survey on Rubric-Guided Reinforcement Learning for Language Models
- The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignment
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks