Update README.md
Browse files
README.md
CHANGED
|
@@ -42,7 +42,7 @@ Total trained query and document pair is 500,000.<br>
|
|
| 42 |
|
| 43 |
### Training Details
|
| 44 |
The overall training process was conducted with reference to snowflake-arctic-2.0.<br>
|
| 45 |
-
In V2, a three-stage training process was introduced as a key component of the overall learning strategy
|
| 46 |
The training process consisted of three stages: Adaptation-training, Pre-training, and Post-training.
|
| 47 |
|
| 48 |
* In the adaptation-training stage, we observed through preliminary experiments that multi-vector Retrieval consistently outperformed standard dense retrieval. To reflect this, we first trained the model using a multi-vector Retrieval objective.
|
|
|
|
| 42 |
|
| 43 |
### Training Details
|
| 44 |
The overall training process was conducted with reference to snowflake-arctic-2.0.<br>
|
| 45 |
+
**In V2, a three-stage training process was introduced as a key component of the overall learning strategy.**
|
| 46 |
The training process consisted of three stages: Adaptation-training, Pre-training, and Post-training.
|
| 47 |
|
| 48 |
* In the adaptation-training stage, we observed through preliminary experiments that multi-vector Retrieval consistently outperformed standard dense retrieval. To reflect this, we first trained the model using a multi-vector Retrieval objective.
|