Unverified50% confidenceFactExact time
GEPA (Genetic-Pareto) is an optimization strategy in DSPy that combines genetic-style search with Pareto-frontier ideas, using execution traces, scoring metrics, and natural language feedback to iteratively rewrite prompts.
1
Sources
50%
Confidence
Long-term
Relevance
10/25/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGEPA (Generalized Evolutionary Prompt Adaptation) is an optimization framework that introduces evolutionary algorithm concepts into AI prompt engineering, screening for the optimal prompt configuration through the generation, evaluation, and elimination of candidate prompts.78% similarUnverifiedGEPA将遗传算法的变异操作与Agent执行轨迹深度绑定，根据轨迹中的具体失败点提出针对性修改而非随机变异71% similarUnverifiedProfile-Guided Optimization (PGO) is a technique that guides compiler optimization by collecting data about a program's actual runtime behavior.64% similarUnverifiedGenVerOpt is a framework that unifies generation, verification, and optimization into a single reasoning pipeline for visual non-verbal reasoning.62% similarUnverifiedDSPy的核心思想是将LLM程序的结构与参数分离，由优化器自动搜索和编译生成最优的提示配置62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1096187API
curl https://kongchang.com/api/v1/knowledge/claims/1096187MCP
get_claim(id=1096187)