Sansa commited on
Commit
be22eca
·
verified ·
1 Parent(s): 36651e2

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +28 -0
README.md ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: unknown
3
+ base_model:
4
+ - apple/DiffuCoder-7B-Instruct
5
+ tags:
6
+ - code
7
+ - text-diffusion-model
8
+ - diffusion large language model
9
+ ---
10
+
11
+ ### DiffuCoder-7B-cpGRPO
12
+
13
+ The DiffuCoder-7B-cpGRPO variant further refines DiffuCoder-Instruct with reinforcement learning via Coupled-GRPO.
14
+
15
+ Training recipe:
16
+
17
+ - Initialized from DiffuCoder-7B-Instruct, post-training with coupled-GRPO on 21K code data (1 epoch).
18
+ - coupled-GRPO significantly improves DiffuCoder's performance on code generation benchmarks (+4.4\% on EvalPlus) and reduces reliance on AR bias during decoding.
19
+
20
+
21
+ #### More details and usage examples:
22
+
23
+ - Paper: [DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation](https://arxiv.org/abs/2506.20639)
24
+
25
+ - GitHub: https://github.com/apple/ml-diffucoder
26
+
27
+ #### Acknowledgement
28
+ To power this HuggingFace model release, we reuse [Dream](https://huggingface.co/Dream-org/Dream-v0-Base-7B)'s modeling architecture and generation utils.