Upload chinese-crypto-importance v1.1

README.md

@@ -17,24 +17,24 @@ metrics:
   - accuracy
   - pearsonr
 model-index:
-  - name: chinese-crypto-importance (v1.0)
+  - name: chinese-crypto-importance (v1.1)
     results:
       - task:
           type: text-classification
           name: News Importance Binning
         metrics:
           - type: mae
-            value: 8.35
+            value: 6.87
             name: MAE
           - type: accuracy
-            value: 70.1%
+            value: 61.8%
             name: Bin Accuracy
           - type: pearsonr
-            value: 0.575
+            value: 0.532
             name: Pearson r
 ---
 
-# Chinese Crypto News Importance Scoring Model | 中文加密货币新闻重要性评分模型 (v1.0)
+# Chinese Crypto News Importance Scoring Model | 中文加密货币新闻重要性评分模型 (v1.1)
 
 ## 模型描述 | Model Description
 
@@ -51,11 +51,11 @@ This model is LoRA fine-tuned from [LocalOptimum/chinese-crypto-sentiment](https
 
 ## 训练数据 | Training Data
 
-- 数据量 | Size: 3364 条中文加密货币新闻样本 | 3364 Chinese crypto news samples
-- 数据来源 | Source: EventAlpha / WatchTower 采集的 3281 条新闻 + 83 条推文 | 3281 news articles + 83 tweets collected via EventAlpha / WatchTower
+- 数据量 | Size: 20286 条中文加密货币新闻样本 | 20286 Chinese crypto news samples
+- 数据来源 | Source: EventAlpha / WatchTower 采集的 19729 条新闻 + 557 条推文 | 19729 news articles + 557 tweets collected via EventAlpha / WatchTower
 - 标注方式 | Labeling: 自动四维评分管线 + 规则修正 | 4-axis automatic scoring pipeline with rule-based cleanup
-- 划分方式 | Split: 随机划分，训练集 2859 / 验证集 505 | Random split with 2859 train and 505 validation samples
-- 平均分数 | Average Score: 41.0
+- 划分方式 | Split: 随机划分，训练集 17243 / 验证集 3043 | Random split with 17243 train and 3043 validation samples
+- 平均分数 | Average Score: 41.7
 
 ### 标注维度 | Scoring Axes
 
@@ -70,10 +70,10 @@ This model is LoRA fine-tuned from [LocalOptimum/chinese-crypto-sentiment](https
 
 | Bin | Score Range | Count | Share | 含义 / Interpretation |
 |---|---:|---:|---:|---|
-| `noise` | 0-25 | 379 | 11.3% | Low-signal, duplicate, digest, or weakly relevant content |
-| `low` | 25-50 | 2272 | 67.5% | Routine updates that rarely move the market on their own |
-| `medium` | 50-75 | 682 | 20.3% | Tradeable developments with meaningful but limited impact |
-| `high` | 75-100 | 31 | 0.9% | Major events that may materially change price or risk appetite |
+| `noise` | 0-25 | 1626 | 8.0% | Low-signal, duplicate, digest, or weakly relevant content |
+| `low` | 25-50 | 14773 | 72.8% | Routine updates that rarely move the market on their own |
+| `medium` | 50-75 | 3840 | 18.9% | Tradeable developments with meaningful but limited impact |
+| `high` | 75-100 | 47 | 0.2% | Major events that may materially change price or risk appetite |
 
 ## 性能指标 | Performance Metrics
 
@@ -81,10 +81,10 @@ This model is LoRA fine-tuned from [LocalOptimum/chinese-crypto-sentiment](https
 
 | 指标 Metric | 数值 Value |
 |---|---:|
-| MAE | 8.35 |
-| Bin Accuracy | 70.1% |
-| Pearson r | 0.575 |
-| Best Epoch | 5 |
+| MAE | 6.87 |
+| Bin Accuracy | 61.8% |
+| Pearson r | 0.532 |
+| Best Epoch | 4 |
 
 ## 分数解释 | Score Interpretation
 
@@ -158,7 +158,7 @@ print(pipe("比特币突破关键阻力位并创下阶段新高"))
 - 基础模型 | Base Model: `LocalOptimum/chinese-crypto-sentiment`
 - 模型结构 | Architecture: BERT backbone + classification head + regression head
 - 最大长度 | Max Length: 256
-- 训练轮数 | Epochs: 10（Early Stopping patience=3，最佳 epoch=5）
+- 训练轮数 | Epochs: 10（Early Stopping patience=3，最佳 epoch=4）
 - 批次大小 | Batch Size: 16
 - 学习率 | Learning Rate: 2e-5
 - LoRA: `r=16`, `alpha=32`, `dropout=0.05`
@@ -203,7 +203,7 @@ Apache-2.0
   author={Onefly},
   year={2026},
   howpublished={\url{https://huggingface.co/LocalOptimum/chinese-crypto-importance}},
-  note={LoRA fine-tuned from LocalOptimum/chinese-crypto-sentiment, 3364 samples, MAE=8.35, BinAcc=70.1%}
+  note={LoRA fine-tuned from LocalOptimum/chinese-crypto-sentiment, 20286 samples, MAE=6.87, BinAcc=61.8%}
 }
 ```
 
@@ -219,7 +219,7 @@ Apache-2.0
 
 - 首个公开的重要性评分模型版本
 - 支持双头输出：连续重要性分数 + 4 档重要性分类
-- 基于 3364 条中文加密货币新闻样本完成训练
-- 当前验证指标：MAE=8.35，Bin Accuracy=70.1%，Pearson r=0.575
+- 基于 20286 条中文加密货币新闻样本完成训练
+- 当前验证指标：MAE=6.87，Bin Accuracy=61.8%，Pearson r=0.532
 
 如有问题或建议，欢迎提 issue 或 PR。

model.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9b8147a6586ee2872183eb9321d549727370dfe477bac46c58c1e8d7549980f
+oid sha256:269cce4fff0f4f5f398bdbd320745f5c21db1ed33826f60bf2b312c86973975e
 size 420517423

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0f95a5d410b802ffad59c5ad34e0984ec546025d90c4da0b317919ff9879307
+oid sha256:20b8f7009f4fcfa23c0a97d9d30353b1608988f7e7f03445c49ee4e89a3bd562
 size 419828528

news_importance_config.json

@@ -10,33 +10,33 @@
     "high"
   ],
   "bin_edges": [
-    25.0,
-    50.0,
-    75.0
+    20.0,
+    35.0,
+    50.0
   ],
   "max_length": 256,
   "metrics": {
-    "epoch": 5,
-    "loss": 0.4292711278413261,
-    "mae": 8.35,
-    "bin_accuracy": 70.1,
-    "pearson_r": 0.575
+    "epoch": 4,
+    "loss": 0.5274954286986246,
+    "mae": 6.87,
+    "bin_accuracy": 61.8,
+    "pearson_r": 0.532
   },
   "dataset": {
-    "samples": 3364,
-    "train_samples": 2859,
-    "eval_samples": 505,
-    "average_score": 41.0,
+    "samples": 20286,
+    "train_samples": 17243,
+    "eval_samples": 3043,
+    "average_score": 41.7,
     "bin_counts": {
-      "noise": 379,
-      "low": 2272,
-      "medium": 682,
-      "high": 31
+      "noise": 1626,
+      "low": 14773,
+      "medium": 3840,
+      "high": 47
     },
     "source_type_counts": {
-      "news": 3281,
-      "tweet": 83
+      "news": 19729,
+      "tweet": 557
     }
   },
-  "version": "v1.0"
+  "version": "v1.1"
 }