Improve model card metadata and add paper link

Hi! I'm Niels from the Hugging Face community team.

I've opened this PR to improve the discoverability of your model by adding `library_name` and `pipeline_tag` metadata. This enables features like the "Use in Transformers" button and ensures the model appears in relevant searches on the Hub.

I've also added a link to your paper [D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation](https://huggingface.co/papers/2603.01780) at the top of the model card.

README.md

@@ -1,14 +1,18 @@
 ---
 license: apache-2.0
+library_name: transformers
+pipeline_tag: text-generation
 tags:
-  - biology
-  - genomics
-  - dna
-  - diffusion
+- biology
+- genomics
+- dna
+- diffusion
 ---
 
 # D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation
 
+This repository contains the weights for **D3LM-scratch**, presented in the paper [D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation](https://huggingface.co/papers/2603.01780).
+
 A masked diffusion language model for **unconditional mammalian DNA sequence generation**, built on the ESM encoder with Rotary Positional Embeddings.
 
 > **Initialization**: trained **from scratch** (random initialization) with masked diffusion objective on mammalian DNA.
@@ -54,7 +58,8 @@ with torch.no_grad():
     outputs = model.diffusion_generate(inputs=input_ids, generation_config=config)
 
 for i, seq in enumerate(outputs.sequences):
-    print(f">{i}\n{tokenizer.decode(seq, skip_special_tokens=True).replace(' ', '')}")
+    print(f">{i}
+{tokenizer.decode(seq, skip_special_tokens=True).replace(' ', '')}")
 ```
 
 ## Generation Parameters
@@ -83,4 +88,4 @@ for i, seq in enumerate(outputs.sequences):
 
 ## License
 
-Apache 2.0
+Apache 2.0