Upload README.md with huggingface_hub

README.md

@@ -15,6 +15,10 @@ configs:
   data_files:
   - split: train
     path: data/et_type.jsonl
+- config_name: difficulty_tag
+  data_files:
+  - split: train
+    path: data/difficulty_tag.jsonl
 ---
 
 # AInsteinBench
@@ -98,6 +102,9 @@ dataset = load_dataset("ByteDance-Seed/AInsteinBench")
 msb_dataset = load_dataset("ByteDance-Seed/AInsteinBench", "msb_type")
 et_dataset = load_dataset("ByteDance-Seed/AInsteinBench", "et_type")
 
+# Load difficulty annotations
+difficulty = load_dataset("ByteDance-Seed/AInsteinBench", "difficulty_tag")
+
 # Access samples
 for sample in dataset['train']:
     print(f"ID: {sample['question_id']}")
@@ -105,6 +112,14 @@ for sample in dataset['train']:
     print(f"Task: {sample['description']}")
 ```
 
+### Difficulty Tags
+
+The `difficulty_tag` subset provides difficulty annotations for each question with two metrics:
+- **`engineeringDifficultyScore`** (1-5): Software engineering complexity
+- **`semanticDepthScore`** (1-5): Domain-specific scientific knowledge required
+
+Match difficulty tags to questions using `repo` and `pr_number` fields.
+
 ### Evaluation
 
 For evaluation scripts and detailed usage, please visit the [AInsteinBench GitHub repository](https://github.com/ByteDance-Seed/AInsteinBench).