Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,39 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: gray
|
| 5 |
colorTo: gray
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: dot company
|
| 3 |
+
emoji: ⚪
|
| 4 |
colorFrom: gray
|
| 5 |
colorTo: gray
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# ● dot company
|
| 11 |
+
|
| 12 |
+
**Researching intelligence from the byte up.**
|
| 13 |
+
|
| 14 |
+
dot company is an independent AI research organization focused on **multimodal, byte-level intelligence**.
|
| 15 |
+
|
| 16 |
+
We explore what happens when models operate closer to raw information, reducing reliance on modality-specific tokenization and hand-designed representations. Our research investigates architectures and training methods capable of learning directly from low-level sequences across text and other modalities.
|
| 17 |
+
|
| 18 |
+
## Research
|
| 19 |
+
|
| 20 |
+
Our work currently centers around:
|
| 21 |
+
|
| 22 |
+
* **Byte-level modeling:** Learning directly from bytes rather than conventional text tokenization.
|
| 23 |
+
* **Multimodal intelligence:** Developing representations and architectures that can generalize across different forms of information.
|
| 24 |
+
* **Small language models:** Studying how much capability can emerge from compact, compute-efficient systems.
|
| 25 |
+
* **Efficient architectures:** Exploring recurrence, parameter sharing, adaptive computation, and alternatives to simply scaling model size.
|
| 26 |
+
* **Representation learning:** Investigating how models can discover useful structure with fewer assumptions about how their inputs should be encoded.
|
| 27 |
+
|
| 28 |
+
## Where we're headed
|
| 29 |
+
|
| 30 |
+
We believe increasingly general models may emerge not only from making systems larger, but from making their foundations **simpler and more universal**.
|
| 31 |
+
|
| 32 |
+
Our long-term direction is toward models that can consume heterogeneous information through a shared low-level interface, learn useful internal representations on their own, and allocate computation intelligently.
|
| 33 |
+
|
| 34 |
+
Instead of building a different foundation for every modality, we're interested in a more fundamental question:
|
| 35 |
+
|
| 36 |
+
> **How little structure can we impose while still allowing intelligence to emerge?**
|
| 37 |
+
|
| 38 |
+
We're starting with bytes.
|
| 39 |
+
|