Add pipeline_tag, library_name, and links to paper/code

README.md

@@ -1,14 +1,21 @@
 ---
-license: mit
 datasets:
 - HuggingFaceTB/smollm-corpus
 language:
 - en
+license: mit
+pipeline_tag: text-generation
+library_name: transformers
 ---
 
 # L2T 500M Shared
 
+This repository contains the **L2T 500M Shared** model, as introduced in the paper [Enhancing Linguistic Competence of Language Models through Pre-training with Language Learning Tasks](https://huggingface.co/papers/2601.03448).
 
+**L2T** (Language Learning Tasks) is a pre-training framework that integrates structured linguistic tasks alongside standard next-token prediction. Inspired by human language acquisition, L2T transforms raw text into structured input-output pairs to provide explicit linguistic stimulation, improving linguistic competence while maintaining competitive performance on general reasoning tasks.
+
+- **Paper:** [Enhancing Linguistic Competence of Language Models through Pre-training with Language Learning Tasks](https://huggingface.co/papers/2601.03448)
+- **Code:** [GitHub - gucci-j/l2t](https://github.com/gucci-j/l2t)
 
 ## How to Get Started with the Model
 Use the code below to get started with the model.
@@ -23,9 +30,8 @@ tokenizer = AutoTokenizer.from_pretrained(
 )
 ```
 
-
 ## Citation
-```
+```bibtex
 @article{yamaguchi2026enhancinglinguisticcompetencelanguage,
       title={Enhancing Linguistic Competence of Language Models through Pre-training with Language Learning Tasks}, 
       author={Atsuki Yamaguchi and Maggie Mi and Nikolaos Aletras},
@@ -37,6 +43,4 @@ tokenizer = AutoTokenizer.from_pretrained(
       journal={arXiv},
       volume={abs/2601.03448}
 }
-```
-
-
+```