docs: rebrand readmes as LLM_QUANT_FACTORY

Author: JIANGJINGZHE-2026

README.md

@@ -1,6 +1,6 @@
 <div align="center">
 
-# AutoAlpha
+# LLM_QUANT_FACTORY
 
 ### 可审计的多智能体 A 股因子研究与组合发现平台
 
@@ -22,7 +22,7 @@
 
 ---
 
-AutoAlpha 的目标不是让大语言模型直接“决定买什么”，而是把模型放在一个可审计、可证伪、
+LLM_QUANT_FACTORY 的目标不是让大语言模型直接“决定买什么”，而是把模型放在一个可审计、可证伪、
 受确定性协议约束的研究系统里。LLM 负责提出机制假设、生成受控表达式和形成结构化研究意见；
 数据时点检查、回测、统计检验、组合权重、风险门禁和策略交付由确定性组件执行。
 
@@ -172,7 +172,7 @@ flowchart LR
 
 ```bash
 git clone https://github.com/khakhasshi/LLM_QUANT_FACTORY.git
-cd MultiFactorAshare
+cd LLM_QUANT_FACTORY
 
 uv sync --frozen --all-groups
 cd AutoAlpha

README_EN.md

@@ -1,6 +1,6 @@
 <div align="center">
 
-# AutoAlpha
+# LLM_QUANT_FACTORY
 
 ### Auditable multi-agent factor research and portfolio discovery for A-shares
 
@@ -24,7 +24,7 @@ search, screening, backtesting, audit trails, and strategy versioning.
 
 ---
 
-AutoAlpha does not ask a language model to directly decide what to buy. Instead, it places LLMs
+LLM_QUANT_FACTORY does not ask a language model to directly decide what to buy. Instead, it places LLMs
 inside an auditable, falsifiable research system governed by deterministic protocols. LLMs propose
 mechanism hypotheses, generate constrained expressions, and produce structured research opinions.
 Data-timing checks, backtests, statistical tests, portfolio weights, risk gates, and strategy
@@ -193,7 +193,7 @@ The system deliberately separates research convenience from production evidence:
 
 ```bash
 git clone https://github.com/khakhasshi/LLM_QUANT_FACTORY.git
-cd MultiFactorAshare
+cd LLM_QUANT_FACTORY
 
 uv sync --frozen --all-groups
 cd AutoAlpha