Update README.md
Browse files
README.md
CHANGED
|
@@ -73,20 +73,7 @@ Unlike traditional supervised fine-tuning (used in ConflLlama), this model uses
|
|
| 73 |
### Training Format
|
| 74 |
- Input: News article describing potential conflict event
|
| 75 |
- Output: Structured XML with reasoning and final category
|
| 76 |
-
|
| 77 |
-
```
|
| 78 |
-
Respond in the following format:
|
| 79 |
-
<reasoning>
|
| 80 |
-
1. Triggers detected: [List any event triggers]
|
| 81 |
-
2. Participants and organizers: [List any actors involved]
|
| 82 |
-
3. Location details: [Specify the location]
|
| 83 |
-
4. Violence assessment: [Indicate if violent or non-violent]
|
| 84 |
-
5. Event category determination: [State and justify the category]
|
| 85 |
-
</reasoning>
|
| 86 |
-
<answer>
|
| 87 |
-
[Final category]
|
| 88 |
-
</answer>
|
| 89 |
-
```
|
| 90 |
|
| 91 |
## Key Mathematical Concepts
|
| 92 |
|
|
|
|
| 73 |
### Training Format
|
| 74 |
- Input: News article describing potential conflict event
|
| 75 |
- Output: Structured XML with reasoning and final category
|
| 76 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
|
| 78 |
## Key Mathematical Concepts
|
| 79 |
|