Spaces:
Configuration error
Configuration error
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,17 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
-
|
| 9 |
-
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# DomainShield
|
| 2 |
+
|
| 3 |
+
**Privacy-first AI for LLM pipelines**
|
| 4 |
+
|
| 5 |
+
DomainShield is a research initiative focused on detecting and protecting sensitive data in large language model pipelines.
|
| 6 |
+
|
| 7 |
+
## What we do
|
| 8 |
+
- PII detection and anonymization
|
| 9 |
+
- Domain-specific entity protection
|
| 10 |
+
- Privacy-preserving AI architectures
|
| 11 |
+
- Multilingual data security
|
| 12 |
+
|
| 13 |
+
## Approach
|
| 14 |
+
We explore prompting, fine-tuning, and modular pipeline designs to enable safe use of third-party LLMs without exposing sensitive information.
|
| 15 |
+
|
| 16 |
+
## Status
|
| 17 |
+
Active research project. Demo and models coming soon.
|