feat: add CSE paper on controlled self-evolution for code optimization.

[itxaiohanglover]

Summary

Add paper "Controlled Self-Evolution for Algorithmic Code Optimization" to the Performance Optimization category.

Paper Details

• Title: Controlled Self-Evolution for Algorithmic Code Optimization
• Authors: Tu Hu, Ronghao Chen, Shuo Zhang, Jianghao Yin, Mou Xiao Feng, Jingping Liu, Shaolei Zhang, Wenqi Jiang,
  Yuqi Fang, Sen Hu, Yi Xu, Huacan Wang
• Venue: arXiv 2026 (2601.07348)

Links

• 📄 Paper
• 💻 GitHub - EvoControl

Overview

CSE (Controlled Self-Evolution) is a framework for algorithmic code optimization that addresses LLM optimization
bottlenecks through:

1. Diversified Initialization: Planning multiple distinct algorithmic strategies (dynamic programming, greedy,
   divide-and-conquer)
2. Intelligent Mutation: Using profiling to identify bottlenecks and "hybridizing" strengths from different code
   versions
3. Hierarchical Memory Bank: Recording failed attempts and general optimization patterns

Compared to Google's AlphaEvolve, CSE focuses more on code engineering optimization from an evolutionary perspective
rather than brute-force search.

Changes

• Modified: data/papers_peformance_optimization.yaml
• Added in reverse chronological order (2026 → 2025)

Checklist

• Follow entry template format
• Place in correct category (Performance Optimization)
• Order by reverse-chronology
• Include arXiv and GitHub links