Update README.md
Browse files
README.md
CHANGED
|
@@ -111,8 +111,10 @@ After the initial filtering, we used dual-consistency filtering to remove potent
|
|
| 111 |
|
| 112 |
During training, we employ a novel curriculum-based hard negative mining strategy to ensure the model learns from challenging examples. We use a softmax-based sampling strategy to progressively sample hard negatives with increasing difficulty over time.
|
| 113 |
|
|
|
|
| 114 |
## Join the Nomic Community
|
| 115 |
|
|
|
|
| 116 |
- Website: [https://nomic.ai](https://nomic.ai)
|
| 117 |
- Twitter: [https://twitter.com/nomic_ai](https://twitter.com/nomic_ai)
|
| 118 |
- Discord: [https://discord.gg/myY5YDR8z8](https://discord.gg/myY5YDR8z8)
|
|
|
|
| 111 |
|
| 112 |
During training, we employ a novel curriculum-based hard negative mining strategy to ensure the model learns from challenging examples. We use a softmax-based sampling strategy to progressively sample hard negatives with increasing difficulty over time.
|
| 113 |
|
| 114 |
+
|
| 115 |
## Join the Nomic Community
|
| 116 |
|
| 117 |
+
- Nomic Embed Ecosystem: [https://www.nomic.ai/embed](https://www.nomic.ai/embed)
|
| 118 |
- Website: [https://nomic.ai](https://nomic.ai)
|
| 119 |
- Twitter: [https://twitter.com/nomic_ai](https://twitter.com/nomic_ai)
|
| 120 |
- Discord: [https://discord.gg/myY5YDR8z8](https://discord.gg/myY5YDR8z8)
|