Add id2label mapping + temperature scaling docs + tags polish

README.md

@@ -9,6 +9,8 @@ tags:
 - chinese
 - bert
 - text-classification
+- binary-classification
+- thesis
 - academic
 base_model: bert-base-chinese
 metrics:
@@ -146,6 +148,7 @@ from transformers import AutoTokenizer, AutoModelForSequenceClassification
 import torch
 
 MODEL_ID = "AnxForever/chinese-ai-detector-bert"
+TEMPERATURE = 0.8165  # Temperature scaling, calibrated on 910 samples (ECE=0.0034)
 
 tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
 model = AutoModelForSequenceClassification.from_pretrained(MODEL_ID)
@@ -156,17 +159,21 @@ inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=256)
 
 with torch.no_grad():
     logits = model(**inputs).logits
-    probs = torch.softmax(logits, dim=-1)[0]
+    # Apply temperature scaling for calibrated confidence
+    probs = torch.softmax(logits / TEMPERATURE, dim=-1)[0]
 
-labels = ["human-written", "AI-generated"]
 pred_idx = int(probs.argmax())
-print(f"{labels[pred_idx]}  (confidence: {probs[pred_idx].item():.2%})")
+label = model.config.id2label[pred_idx]   # "human-written" or "AI-generated"
+print(f"{label}  (confidence: {probs[pred_idx].item():.2%})")
 ```
 
 ### Label mapping
 - `0` → human-written (人类撰写)
 - `1` → AI-generated (AI 生成)
 
+> **Note on Temperature Scaling**: `T = 0.8165` was calibrated on a held-out 910-sample set
+> and brings ECE from 0.0121 down to **0.0034**. For uncalibrated probabilities, set `TEMPERATURE = 1.0`.
+
 ---
 
 ## 🎯 技术创新 / Contributions

config.json

@@ -9,8 +9,16 @@
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
+  "id2label": {
+    "0": "human-written",
+    "1": "AI-generated"
+  },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
+  "label2id": {
+    "AI-generated": 1,
+    "human-written": 0
+  },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",