evolutionary-optimization

Tag

Cards List
#evolutionary-optimization

SkillNEO: Non-Elitist Evolutionary Optimization of Agent Skills

arXiv cs.AI ↗ · 2026-10-16 Cached

SkillNEO 提出非精英进化优化方法来改进 LLM agent 技能,通过允许低奖励版本继续迭代并分析版本间差异来保留有效步骤、修复失败,使 GPT-5.4 Mini 在 86 项 SkillsBench 任务上平均奖励从 49.1% 提升至 66.4%,且改进后的技能还能提升更强模型的表现。

0 favorites 0 likes
#evolutionary-optimization

Self-Evolving Agent Harness

Hugging Face Daily Papers ↗ · 2026-10-13 Cached

提出一个自我进化代理框架(Harness RSI),在保持模型权重固定的情况下,通过反馈驱动的修订与进化优化来持续改进代理的软件脚手架(提示、记忆、工作流和可执行组件),并借助持续 PAC 学习给出泛化与保留的理论保证。

0 favorites 0 likes
#evolutionary-optimization

Meta-Reflective Evolution Effectively Improves Production AI Systems

arXiv cs.LG ↗ · 2026-10-12 Cached

论文提出 CoCoEvolve(CortexCodeEvolve),一种基于"元反思"的进化优化器,通过从完整搜索历史中检索证据并主动探测目标环境来改进生产级 AI 系统的优化,在 DABStep、data-eng-bench 等多个基准上超越现有基线。

0 favorites 0 likes
#evolutionary-optimization

VIEWCRAFT: Evolving What Multimodal Agents See and How They Read It

Hugging Face Daily Papers ↗ · 2026-10-11 Cached

VIEWCRAFT 通过联合进化多模态代理的证据界面与读取流程,在八个基准上平均提升 6.1 分任务得分,并以更少的输入 token 实现跨执行器迁移。

0 favorites 0 likes
#evolutionary-optimization

HeurEvo: Agentic Evolution of Hybrid Solver-Augmented Heuristics for Time-Critical Mathematical Optimization

arXiv cs.CL ↗ · 2026-09-30 Cached

HeurEvo introduces an automated plan–code–component co-evolution framework in which AI agents jointly evolve algorithmic structure, executable implementations, and a shared pool of reusable components for hybrid solver-augmented heuristics. Across combinatorial optimization benchmarks and MIPLIB instances, it matches or surpasses state-of-the-art solvers under tight runtime budgets and sets new results on problems such as hexagon packing.

0 favorites 0 likes
#evolutionary-optimization

Rank-Reliable Teacher-Guided Fitness Approximation for Expensive Evolutionary Optimization: A TinyML Architecture Search Study

arXiv cs.AI ↗ · 2026-09-28 Cached

The paper introduces TGL-NSGA-II, a teacher-guided fitness approximation framework for expensive evolutionary optimization in constrained TinyML neural architecture search, achieving improved efficiency and reliability over standard methods.

0 favorites 0 likes
#evolutionary-optimization

GraphSkillEvo: Evolutionary Optimization of Graph-Structured Agent Skills

Hugging Face Daily Papers ↗ · 2026-09-18 Cached

GraphSkillEvo is an evolutionary optimization framework that represents agent skills as graph-structured artifacts to improve LLM performance, outperforming baselines on multiple benchmarks.

0 favorites 0 likes
#evolutionary-optimization

@dair_ai: // Evolving Meta-Skill for Multi-Agent Systems // Can a multi-agent system get better at orchestration without touching…

X AI KOLs Following ↗ · 2026-06-20 Cached

Skill-MAS introduces a method for evolving meta-skills in multi-agent systems to improve orchestration without modifying model weights, achieving transferable performance gains across tasks and LLMs.

0 favorites 0 likes
#evolutionary-optimization

Residual-Space Evolutionary Optimization via Flow-based Generative Models

arXiv cs.AI ↗ · 2026-06-20 Cached

Introduces a framework combining flow-based generative editing with evolutionary algorithms to perform optimization in residual space, enabling controllable data editing with non-differentiable objectives. Validated on MorphoMNIST and crystal data.

0 favorites 0 likes
#evolutionary-optimization

Environment-Grounded Automated Prompt Optimization for LLM Game Agents

arXiv cs.CL ↗ · 2026-06-17 Cached

Introduces an automated prompt optimization framework for LLM game agents that decomposes the observation-to-action pipeline into two agents and iteratively refines prompts via an evolutionary loop guided by environment returns. Evaluated on BabyAI tasks, it significantly improves success rates (e.g., from 0% to 72.5% on PutNext) without updating model weights.

0 favorites 0 likes
#evolutionary-optimization

@Sumanth_077: Stop testing and rewriting prompts manually! Most teams run evals, look at failures, guess what's wrong, rewrite the pr…

X AI KOLs Timeline ↗ · 2026-05-14 Cached

DeepEval introduces an evolutionary optimization method for prompts using genetic algorithms, allowing automatic rewriting based on eval feedback and multi-objective optimization.

0 favorites 0 likes
#evolutionary-optimization

Evolutionary fine tuning of quantized convolution-based deep learning models

arXiv cs.LG ↗ · 2026-05-08 Cached

This paper proposes a neuroevolution-based fine-tuning method to improve the accuracy of quantized deep learning models, showing that nearest-neighbor rounding alone is suboptimal and that evolutionary mutation of weights can yield better results on architectures like VGG and ResNet.

0 favorites 0 likes
← Back to home

Submit Feedback