Add pipeline tag and sample usage

#1
by nielsr HF Staff - opened
Files changed (1) hide show
  1. README.md +22 -0
README.md CHANGED
@@ -1,5 +1,6 @@
1
  ---
2
  license: mit
 
3
  ---
4
 
5
  # Mixture-Outerlinks
@@ -9,6 +10,27 @@ license: mit
9
  We introduce RecursiveMAS, a multi-agent framework that scales agent collaboration through latent-space recursion.
10
  RecursiveMAS treats a multi-agent system as a unified recursive computation, where heterogeneous agents iteratively exchange, refine, and evolve their latent states across recursion rounds. In the Mixture-Style setting, domain-specialized agents collaborate with the Summarizer Agent through Outer RecursiveLink modules for integrating specialized reasoning into the final response.
11
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
  ## Model Details
13
 
14
  | Item | Description |
 
1
  ---
2
  license: mit
3
+ pipeline_tag: text-generation
4
  ---
5
 
6
  # Mixture-Outerlinks
 
10
  We introduce RecursiveMAS, a multi-agent framework that scales agent collaboration through latent-space recursion.
11
  RecursiveMAS treats a multi-agent system as a unified recursive computation, where heterogeneous agents iteratively exchange, refine, and evolve their latent states across recursion rounds. In the Mixture-Style setting, domain-specialized agents collaborate with the Summarizer Agent through Outer RecursiveLink modules for integrating specialized reasoning into the final response.
12
 
13
+ ## Usage
14
+
15
+ This repository provides the **Outer RecursiveLink modules** for the **Mixture-Style** configuration of **RecursiveMAS**. You can load the full multi-agent system using the following snippet (requires the [GitHub repository](https://github.com/RecursiveMAS/RecursiveMAS) code):
16
+
17
+ ```python
18
+ from system_loader import load_mas_system
19
+
20
+ # Load the full Mixture-Style MAS pipeline
21
+ mas = load_mas_system(
22
+ style="mixture",
23
+ device="cuda",
24
+ trust_remote_code=True,
25
+ )
26
+
27
+ # Access individual agents and their RecursiveLink-connected components
28
+ math_agent = mas.agents["Math"].model
29
+ code_agent = mas.agents["Code"].model
30
+ science_agent = mas.agents["Science"].model
31
+ summarizer = mas.agents["Summarizer"].model
32
+ ```
33
+
34
  ## Model Details
35
 
36
  | Item | Description |