Update README.md
Browse files
README.md
CHANGED
|
@@ -2,6 +2,10 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
```
|
| 6 |
Standard: x_t → UNet(x_t, t, c) → v_pred
|
| 7 |
↑ full 16384-dim path
|
|
@@ -13,6 +17,7 @@ GLFM: x_t → encoder → project to S^15 → triangulate against anchors
|
|
| 13 |
conditioned_generator(address, t, c) → v_pred
|
| 14 |
```
|
| 15 |
|
|
|
|
| 16 |
This one is going to attempt a new form of prediction based on velocity flow matching using spherical representation.
|
| 17 |
|
| 18 |
The alignment for the early prototype is essentially going to be euler discreet ODE flow matching, and it will be more than enough for now.
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
|
| 5 |
+
# Geometric Lookup Flow Matching (GLFM)
|
| 6 |
+
|
| 7 |
+
The prototype formula for the next experiment.
|
| 8 |
+
|
| 9 |
```
|
| 10 |
Standard: x_t → UNet(x_t, t, c) → v_pred
|
| 11 |
↑ full 16384-dim path
|
|
|
|
| 17 |
conditioned_generator(address, t, c) → v_pred
|
| 18 |
```
|
| 19 |
|
| 20 |
+
# Prelim
|
| 21 |
This one is going to attempt a new form of prediction based on velocity flow matching using spherical representation.
|
| 22 |
|
| 23 |
The alignment for the early prototype is essentially going to be euler discreet ODE flow matching, and it will be more than enough for now.
|