Improve model card: Add metadata and update paper/GitHub links (#1)
Browse files- Improve model card: Add metadata and update paper/GitHub links (25638fcd212e5609a54ddf55fb152f0b59b1799f)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,9 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# CDLM-LLaDA LoRA adapter for LLaDA-8B-Instruct
|
| 2 |
|
| 3 |
This repository hosts the LoRA adapter for the LLaDA-8B-Instruct diffusion LLM (dLLM), produced with the CDLM (Consistency Diffusion Language Models) method. CDLM integrates consistency modeling and a block-wise causal attention mask so the student model becomes fully KV-cache compatible while retaining the strong local bidirectional modeling within each block. In practice, the adapter enables significantly faster inference with competitive quality.
|
| 4 |
|
| 5 |
-
- GitHub: https://github.com/
|
| 6 |
-
- Paper:
|
| 7 |
|
| 8 |
|
| 9 |
## Model details
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
pipeline_tag: text-generation
|
| 4 |
+
library_name: peft
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
# CDLM-LLaDA LoRA adapter for LLaDA-8B-Instruct
|
| 8 |
|
| 9 |
This repository hosts the LoRA adapter for the LLaDA-8B-Instruct diffusion LLM (dLLM), produced with the CDLM (Consistency Diffusion Language Models) method. CDLM integrates consistency modeling and a block-wise causal attention mask so the student model becomes fully KV-cache compatible while retaining the strong local bidirectional modeling within each block. In practice, the adapter enables significantly faster inference with competitive quality.
|
| 10 |
|
| 11 |
+
- GitHub: https://github.com/SqueezeAILab/CDLM
|
| 12 |
+
- Paper: [CDLM: Consistency Diffusion Language Models For Faster Sampling](https://huggingface.co/papers/2511.19269)
|
| 13 |
|
| 14 |
|
| 15 |
## Model details
|