Update README.md
Browse files
README.md
CHANGED
|
@@ -20,13 +20,26 @@ We present $\color{goldenrod}{SPHINX}$, a versatile multi-modal large language m
|
|
| 20 |
|
| 21 |
- **Domain Mix.** For data from real-world and synthetic domains, we mix the weights of two domain-specific models for complementarity.
|
| 22 |
|
| 23 |
-
<p align="
|
| 24 |
-
<img src="
|
| 25 |
</p>
|
| 26 |
-
<p align="
|
| 27 |
-
<img src="
|
| 28 |
</p>
|
| 29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
|
| 31 |
## Inference
|
| 32 |
This section provides a step-by-step guide for hosting a local SPHINX demo. If you're already familiar with the LLAMA2-Accessory toolkit, note that hosting a SPHINX demo follows the same pipeline as hosting demos for the other models supported by LLAMA2-Accessory.
|
|
|
|
| 20 |
|
| 21 |
- **Domain Mix.** For data from real-world and synthetic domains, we mix the weights of two domain-specific models for complementarity.
|
| 22 |
|
| 23 |
+
<p align="left">
|
| 24 |
+
<img src="figs/pipeline1.png"/ width="60%"> <br>
|
| 25 |
</p>
|
| 26 |
+
<p align="left">
|
| 27 |
+
<img src="figs/pipeline2.png"/ width="60%"> <br>
|
| 28 |
</p>
|
| 29 |
|
| 30 |
+
## Result
|
| 31 |
+
<p align="left">
|
| 32 |
+
<img src="figs/table1.png"/ width="50%"> <br>
|
| 33 |
+
</p>
|
| 34 |
+
<p align="left">
|
| 35 |
+
<img src="figs/table2.png"/ width="50%"> <br>
|
| 36 |
+
</p>
|
| 37 |
+
<p align="left">
|
| 38 |
+
<img src="figs/table3.png"/ width="50%"> <br>
|
| 39 |
+
</p>
|
| 40 |
+
<p align="left">
|
| 41 |
+
<img src="figs/table4.png"/ width="50%"> <br>
|
| 42 |
+
</p>
|
| 43 |
|
| 44 |
## Inference
|
| 45 |
This section provides a step-by-step guide for hosting a local SPHINX demo. If you're already familiar with the LLAMA2-Accessory toolkit, note that hosting a SPHINX demo follows the same pipeline as hosting demos for the other models supported by LLAMA2-Accessory.
|