1
Spicing up Genetic Netlist Generation with LLMs
用大模型为遗传网表生成注入新配方，芯片设计自动化迎来AI调味时刻。
arXiv:2608.23317v1 Announce Type: cross Abstract: Analog circuit topology synthesis remains challenging because useful designs occupy a tiny fraction …
2
Show HN: Hana JIT – LLVM-backed Python JIT with genetic-algorithm superoptimizer
基于LLVM后端的Python JIT编译器，引入遗传算法超优化器，性能可达CPython的10-100倍
I wanted a replacement for Numba that supports other GPU types and FPGA, I need help in devlopment and testing especially in AMD targets and FPGA. Tha…
3
GAS-Leak-LLM: Genetic Algorithm-Based Suffix Optimization for Black-Box LLM Jailbreaking
用遗传算法自动搜索黑盒大模型的越狱后缀，突破安全护栏的新攻击方法，副作用值得警惕。
arXiv:2606.15788v1 Announce Type: cross Abstract: Large Language Models (LLMs) constitute pivotal components within the AI-dominated information techn…
4
Show HN: Stochastc Unit Commmitment Using Genetic Algorithm
用遗传算法解决电力市场的不确定性决策，一个跳出线性规划的仿真引擎，Julia实现开源。
Hey everyone, in the last weeks I've been thinking about building a power market simulation engine that is not based on any linear or mixed-integer pr…
5
Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolution
LLM驱动的程序演化中，突变缺乏变异导致收敛问题，揭示算法局限性。
arXiv:2606.05408v1 Announce Type: new Abstract: When an LLM repeatedly mutates a program, does it explore new forms or circle back to the same ones? W…
6
The End of Manual Prompt Engineering: How Genetic-Pareto Prompt Evolution (GEPA) Self-Optimizes AI Agents
告别手动调优！GEPA算法融合遗传进化与帕累托优化，让AI Agent自动进化提示词，效率飞升。
If you have spent any time building production-grade LLM applications, you know the dirty secret of the industry: prompt engineering is a vibe-based u…
7
Inducing Overthink: Hierarchical Genetic Algorithm-based DoS Attack on Black-Box Large Language Reasoning Models
新研究揭示如何用分层遗传算法对黑盒大语言推理模型实施DoS攻击，引发对AI安全的新思考。
arXiv:2605.13338v2 Announce Type: replace-cross Abstract: Large Reasoning Models (LRMs) are increasingly integrated into systems requiring reliable mu…
8
GEAR: Genetic AutoResearch for Agentic Code Evolution
论文提出一种遗传式自动研究方法，让AI代理在代码进化中保留多样性的候选方案，避免陷入局部最优。
arXiv:2605.13874v1 Announce Type: cross Abstract: Autonomous research agents can already run machine learning experiments without human supervision, b…