Name the main decision policies FinQ and FinL

Model_Card_EN.md

@@ -3,7 +3,7 @@ language:
 - en
 license: other
 license_name: s2-mas-controlled-research-v1.2
-license_link: https://huggingface.co/ya9919/fin-laya-mlx-6-rounds/blob/main/LICENSE
+license_link: https://huggingface.co/ya9919/FinL-mlx/blob/main/LICENSE
 base_model: convaiinnovations/laya
 base_model_relation: finetune
 tags:
@@ -12,17 +12,21 @@ tags:
 - decision-model
 ---
 
-# Fin-Laya 6-rounds · MLX
+# FinL-mlx · full six-round checkpoint (MLX FP16)
+
+## Model naming and experiment labels
+
+**FinL-mlx — full English-base six-round FinL checkpoint, MLX FP16** is the primary MLX runtime. The native Laya/PyTorch release is [**FinL**](https://huggingface.co/ya9919/FinL). The separately evaluated bilingual deployment checkpoints are named **FinL-bilingual** and **FinL-bilingual-mlx**.
 
 ## Project background
 
-S²-MAS is a society-scale multi-agent simulation framework for studying how heterogeneous household decisions become population outcomes under changing conditions. It connects a constructed resource of **500 million household personas**, reusable scenarios, trainable decision models and stateful world updates. Fin-Laya supplies a reusable decision policy: it scores the complete alternatives for a household state and question, allowing the same policy interface to serve different tasks.
+S²-MAS is a society-scale multi-agent simulation framework for studying how heterogeneous household decisions become population outcomes under changing conditions. It connects a constructed resource of **500 million household personas**, reusable scenarios, trainable decision models and stateful world updates. FinL supplies a reusable decision policy: it scores the complete alternatives for a household state and question, allowing the same policy interface to serve different tasks.
 
 The manuscript's paired-condition application evaluates **12 million distinct synthetic household agents and 24 million choices**. A 24-month extension follows 10,000 households through changing income, expenses and bank offers. The applications show how household resources shape consumption responses: persistent income loss draws down reserves and delays purchases, while sustained income gains support both cash reserves and purchases. Repayment feedback affects later offers. These findings connect individual choices to consumption, cash and payment paths over time.
 
 ## Repository role and training
 
-This is the **MLX FP16 runtime of the English-base Fin-Laya 6-round checkpoint**, paired with the [native release](https://huggingface.co/ya9919/fin-laya-6-rounds). The approximately 421M-parameter Laya encoder and option-scoring head continue the 3-round financial checkpoint through SFT and RLCD over **2,925,263 training examples per language**, with expanded supervision from real-survey data. This is the largest disclosed domain post-training set among the financial decision models compared in the manuscript.
+This is the **MLX FP16 runtime of the English-base FinL 6-round checkpoint**, paired with the [native release](https://huggingface.co/ya9919/FinL). The approximately 421M-parameter Laya encoder and option-scoring head continue the 3-round financial checkpoint through SFT and RLCD over **2,925,263 training examples per language**, with expanded supervision from real-survey data. This is the largest disclosed domain post-training set among the financial decision models compared in the manuscript.
 
 The manuscript's 12M-agent purchase-funding application and 24-month extension use this frozen checkpoint in MLX CUDA FP16.
 
@@ -39,7 +43,7 @@ On the same English shared test, Jev 1.13 scores 65.2584% accuracy and 37.4793%
 
 ## Deployment timing
 
-The manuscript's single-request timing uses the [bilingual Fin-Laya checkpoint](https://huggingface.co/ya9919/fin-laya-multilingual-mlx) on one NVIDIA A100 PCIe 40GB in FP16, with batch size and concurrency one and three fixed Chinese cases. Its card and Methods M9 report latency, throughput and the Jev comparison.
+The manuscript's single-request timing uses the [FinL-bilingual checkpoint](https://huggingface.co/ya9919/fin-laya-multilingual-mlx) on one NVIDIA A100 PCIe 40GB in FP16, with batch size and concurrency one and three fixed Chinese cases. Its card and Methods M9 report latency, throughput and the Jev comparison.
 
 ## Model files and inference
 

Model_Card_ZH.md

@@ -3,7 +3,7 @@ language:
 - en
 license: other
 license_name: s2-mas-controlled-research-v1.2
-license_link: https://huggingface.co/ya9919/fin-laya-mlx-6-rounds/blob/main/LICENSE
+license_link: https://huggingface.co/ya9919/FinL-mlx/blob/main/LICENSE
 base_model: convaiinnovations/laya
 base_model_relation: finetune
 tags:
@@ -12,17 +12,21 @@ tags:
 - decision-model
 ---
 
-# Fin-Laya 6-rounds · MLX
+# FinL-mlx · full six-round checkpoint (MLX FP16)
+
+## 模型命名与实验标签
+
+**FinL-mlx——英语基座完整六轮 FinL checkpoint，MLX FP16 格式**是主要的 MLX 运行版本。原生 Laya/PyTorch 版本为 [**FinL**](https://huggingface.co/ya9919/FinL)。另行评测的双语部署 checkpoint 分别命名为 **FinL-bilingual** 与 **FinL-bilingual-mlx**。
 
 ## 项目背景
 
-S²-MAS 是面向社会规模的多智能体仿真框架，用于研究异质家庭在条件变化下的决策如何汇总为人口层面的结果。它将**5 亿条构造家庭 persona 资源**、可复用场景、可训练决策模型与持续更新的世界状态连接起来。Fin-Laya 在其中提供可复用的决策策略：根据家庭状态、问题和完整候选选项进行评分，使同一决策接口能够服务不同任务。
+S²-MAS 是面向社会规模的多智能体仿真框架，用于研究异质家庭在条件变化下的决策如何汇总为人口层面的结果。它将**5 亿条构造家庭 persona 资源**、可复用场景、可训练决策模型与持续更新的世界状态连接起来。FinL 在其中提供可复用的决策策略：根据家庭状态、问题和完整候选选项进行评分，使同一决策接口能够服务不同任务。
 
 论文中的配对条件应用实际评估了 **1,200 万个独立合成家庭 agent、共 2,400 万次选择**；24 个月扩展则跟踪 10,000 户家庭在收入、必要支出与银行 offer 变化下的状态。应用结果说明家庭资源如何改变消费响应：持续降收消耗现金储备并推迟购置，持续增收则同时支持现金储备与购置；偿付反馈进一步影响后续 offer，从而将个体选择接到消费、现金与支付的时间路径。
 
 ## 本仓库的模型角色与训练
 
-本仓库是**英语基座 Fin-Laya 6-round checkpoint 的 MLX FP16 运行版本**，与[原生版本](https://huggingface.co/ya9919/fin-laya-6-rounds)配套。约 421M 参数的 Laya encoder 与选项评分 head 在 3-round 金融 checkpoint 基础上，以**每语言 2,925,263 条训练样本**继续进行 SFT 与 RLCD，使用来自 real-survey data 的扩展监督。在论文比较且披露训练规模的金融决策模型中，这是最大的领域后训练样本集。
+本仓库是**英语基座 FinL 6-round checkpoint 的 MLX FP16 运行版本**，与[原生版本](https://huggingface.co/ya9919/FinL)配套。约 421M 参数的 Laya encoder 与选项评分 head 在 3-round 金融 checkpoint 基础上，以**每语言 2,925,263 条训练样本**继续进行 SFT 与 RLCD，使用来自 real-survey data 的扩展监督。在论文比较且披露训练规模的金融决策模型中，这是最大的领域后训练样本集。
 
 论文的 1,200 万 agent 购置融资应用和 24 个月扩展均使用此冻结 checkpoint 的 MLX CUDA FP16 格式。
 
@@ -39,7 +43,7 @@ Accuracy 对 QA 记录等权计分；question-macro F1 先在每个问题版本
 
 ## 部署速度
 
-论文的单请求速度测量使用[双语 Fin-Laya checkpoint](https://huggingface.co/ya9919/fin-laya-multilingual-mlx)，条件为一张 NVIDIA A100 PCIe 40GB、FP16、batch size 与并发均为 1、三个固定中文案例。对应延迟、吞吐量与 Jev 比较见该版本卡片及 Methods M9。
+论文的单请求速度测量使用[FinL-bilingual checkpoint](https://huggingface.co/ya9919/fin-laya-multilingual-mlx)，条件为一张 NVIDIA A100 PCIe 40GB、FP16、batch size 与并发均为 1、三个固定中文案例。对应延迟、吞吐量与 Jev 比较见该版本卡片及 Methods M9。
 
 ## 模型文件与推理
 

README.md

@@ -3,7 +3,7 @@ language:
 - en
 license: other
 license_name: s2-mas-controlled-research-v1.2
-license_link: https://huggingface.co/ya9919/fin-laya-mlx-6-rounds/blob/main/LICENSE
+license_link: https://huggingface.co/ya9919/FinL-mlx/blob/main/LICENSE
 base_model: convaiinnovations/laya
 base_model_relation: finetune
 tags:
@@ -12,17 +12,21 @@ tags:
 - decision-model
 ---
 
-# Fin-Laya 6-rounds · MLX
+# FinL-mlx · full six-round checkpoint (MLX FP16)
+
+## Model naming and experiment labels
+
+**FinL-mlx — full English-base six-round FinL checkpoint, MLX FP16** is the primary MLX runtime. The native Laya/PyTorch release is [**FinL**](https://huggingface.co/ya9919/FinL). The separately evaluated bilingual deployment checkpoints are named **FinL-bilingual** and **FinL-bilingual-mlx**.
 
 ## Project background
 
-S²-MAS is a society-scale multi-agent simulation framework for studying how heterogeneous household decisions become population outcomes under changing conditions. It connects a constructed resource of **500 million household personas**, reusable scenarios, trainable decision models and stateful world updates. Fin-Laya supplies a reusable decision policy: it scores the complete alternatives for a household state and question, allowing the same policy interface to serve different tasks.
+S²-MAS is a society-scale multi-agent simulation framework for studying how heterogeneous household decisions become population outcomes under changing conditions. It connects a constructed resource of **500 million household personas**, reusable scenarios, trainable decision models and stateful world updates. FinL supplies a reusable decision policy: it scores the complete alternatives for a household state and question, allowing the same policy interface to serve different tasks.
 
 The manuscript's paired-condition application evaluates **12 million distinct synthetic household agents and 24 million choices**. A 24-month extension follows 10,000 households through changing income, expenses and bank offers. The applications show how household resources shape consumption responses: persistent income loss draws down reserves and delays purchases, while sustained income gains support both cash reserves and purchases. Repayment feedback affects later offers. These findings connect individual choices to consumption, cash and payment paths over time.
 
 ## Repository role and training
 
-This is the **MLX FP16 runtime of the English-base Fin-Laya 6-round checkpoint**, paired with the [native release](https://huggingface.co/ya9919/fin-laya-6-rounds). The approximately 421M-parameter Laya encoder and option-scoring head continue the 3-round financial checkpoint through SFT and RLCD over **2,925,263 training examples per language**, with expanded supervision from real-survey data. This is the largest disclosed domain post-training set among the financial decision models compared in the manuscript.
+This is the **MLX FP16 runtime of the English-base FinL 6-round checkpoint**, paired with the [native release](https://huggingface.co/ya9919/FinL). The approximately 421M-parameter Laya encoder and option-scoring head continue the 3-round financial checkpoint through SFT and RLCD over **2,925,263 training examples per language**, with expanded supervision from real-survey data. This is the largest disclosed domain post-training set among the financial decision models compared in the manuscript.
 
 The manuscript's 12M-agent purchase-funding application and 24-month extension use this frozen checkpoint in MLX CUDA FP16.
 
@@ -39,7 +43,7 @@ On the same English shared test, Jev 1.13 scores 65.2584% accuracy and 37.4793%
 
 ## Deployment timing
 
-The manuscript's single-request timing uses the [bilingual Fin-Laya checkpoint](https://huggingface.co/ya9919/fin-laya-multilingual-mlx) on one NVIDIA A100 PCIe 40GB in FP16, with batch size and concurrency one and three fixed Chinese cases. Its card and Methods M9 report latency, throughput and the Jev comparison.
+The manuscript's single-request timing uses the [FinL-bilingual checkpoint](https://huggingface.co/ya9919/fin-laya-multilingual-mlx) on one NVIDIA A100 PCIe 40GB in FP16, with batch size and concurrency one and three fixed Chinese cases. Its card and Methods M9 report latency, throughput and the Jev comparison.
 
 ## Model files and inference