# Escher-Loop: Mutual Evolution by Closed-Loop Self-Referential Optimization

arXiv: https://arxiv.org/abs/2604.23472 · HTML: https://mlsanity.com/paper/2604.23472/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 46 references that are in the map.
- First posted on arXiv 25 Apr 2026.

## Builds on (most cited first, 18 of 18)

- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [Illuminating search spaces by mapping elites](https://mlsanity.com/paper/1504.04909/) (2015)
- [Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions](https://mlsanity.com/paper/2306.02224/) (2023)
- [Large Language Model Agent: A Survey on Methodology, Applications and Challenges](https://mlsanity.com/paper/2503.21460/) (2025)
- [Learning to Discover at Test Time](https://mlsanity.com/paper/2601.16175/) (2026)
- [ThetaEvolve: Test-time Learning on Open Problems](https://mlsanity.com/paper/2511.23473/) (2025)
- [AdaEvolve: Adaptive LLM Driven Zeroth-Order Optimization](https://mlsanity.com/paper/2602.20133/) (2026)
- [EvoX: Meta-Evolution for Automated Discovery](https://mlsanity.com/paper/2602.23413/) (2026)
- [Hyperagents](https://mlsanity.com/paper/2603.19461/) (2026)
- [CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery](https://mlsanity.com/paper/2604.01658/) (2026)
- [Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing](https://mlsanity.com/paper/2602.04837/) (2026)
- [CoEvolve: Training LLM Agents via Agent-Data Mutual Evolution](https://mlsanity.com/paper/2604.15840/) (2026)
- [Structured Scaling of AI Discovery Across Diverse Scientific Domains](https://mlsanity.com/paper/2604.19341/) (2026)
- [Eliminating Meta Optimization Through Self-Referential Meta Learning](https://mlsanity.com/paper/2212.14392/) (2022)
- [Better LLM Reasoning via Dual-Play](https://mlsanity.com/paper/2511.11881/) (2025)
- [TerraLingua: Emergence and Analysis of Open-endedness in LLM Ecologies](https://mlsanity.com/paper/2603.16910/) (2026)
- [TodoEvolve: Learning to Architect Agent Planning Systems](https://mlsanity.com/paper/2602.07839/) (2026)
- [AgentGA: Evolving Code Solutions in Agent-Seed Space](https://mlsanity.com/paper/2604.14655/) (2026)

## Built on by (most cited first, 8 of 8)

- [Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops](https://mlsanity.com/paper/2607.07663/) (2026)
- [Evolving Ensemble of Agents](https://mlsanity.com/paper/2605.09018/) (2026)
- [Agentic Symbolic Search: Characterizing PDEs Beyond Hand-crafted Expressions, Meshes, and Neural Networks](https://mlsanity.com/paper/2606.20467/) (2026)
- [MetaRSI / RSI2: A Meta-Recursive Self-Improving System for Recursive Self-Improving Systems Themselves](https://mlsanity.com/paper/2609.06396/) (2026)
- [Algorithmic algorithm development with LLMs: A Case Study on LLM-Usage for Contraction Order Optimization in Tensor Networks](https://mlsanity.com/paper/2606.01975/) (2026)
- [ArtMine: Discovering and Formalizing Artistic Processes](https://mlsanity.com/paper/2607.08331/) (2026)
- [From Existing Methods to New Research Ideas: Two Perspectives on Recursive Self-Improvement](https://mlsanity.com/paper/2610.14625/) (2026)
- [Heterogeneous Agentic Coevolution](https://mlsanity.com/paper/2610.16078/) (2026)

## Abstract (arXiv)

While recent autonomous agents demonstrate impressive capabilities, they predominantly rely on manually scripted workflows and handcrafted heuristics, inherently limiting their potential for open-ended improvement. To address this, we propose Escher-Loop, a fully closed-loop framework that operationalizes the mutual evolution of two distinct populations: Task Agents that solve concrete problems, and Optimizer Agents that recursively refine both the task agents and themselves. To sustain this self-referential evolution, we propose a dynamic benchmarking mechanism that seamlessly reuses the empirical scores of newly generated task agents as relative win-loss signals to update optimizers' scores. This mechanism leverages the evolution of task agents as an inherent signal to drive the evaluation and refinement of optimizers without additional overhead. Empirical evaluations on mathematical optimization problems demonstrate that Escher-Loop effectively pushes past the performance ceilings of static baselines, achieving the highest absolute peak performance across all evaluated tasks under matched compute. Remarkably, we observe that the optimizer agents dynamically adapt their strategies to match the shifting demands of high-performing task agents, which explains the system's continuous improvement and superior late-stage performance.