Improve model card: Add pipeline tag, library name, paper and GitHub links

This PR enhances the model card for `F2LLM-4B` by:
- Adding `pipeline_tag: feature-extraction` to correctly categorize the model's functionality on the Hub.
- Including `library_name: transformers` to enable the automated "How to use" widget, as evidenced by the existing usage snippet.
- Adding direct links to the paper ([F2LLM Technical Report: Matching SOTA Embedding Performance with 6 Million Open-Source Data](https://huggingface.co/papers/2510.02294)) and the GitHub repository for better visibility and easier access.

These changes will improve the model's discoverability and usability on the Hugging Face Hub.

README.md

@@ -1,13 +1,21 @@
 ---
-license: apache-2.0
+base_model:
+- Qwen/Qwen3-4B
 datasets:
 - codefuse-ai/F2LLM
 language:
 - en
-base_model:
-- Qwen/Qwen3-4B
+license: apache-2.0
+pipeline_tag: feature-extraction
+library_name: transformers
 ---
 
+# F2LLM-4B: Matching SOTA Embedding Performance with 6 Million Open-Source Data
+
+This model is a part of the F2LLM family, presented in the paper [F2LLM Technical Report: Matching SOTA Embedding Performance with 6 Million Open-Source Data](https://huggingface.co/papers/2510.02294).
+
+**Code**: [https://github.com/codefuse-ai/CodeFuse-Embeddings/tree/main/F2LLM](https://github.com/codefuse-ai/CodeFuse-Embeddings/tree/main/F2LLM)
+
 F2LLMs (Foundation to Feature Large Language Models) are foundation models directly finetuned on 6 million high-quality query-document pairs (available in [codefuse-ai/F2LLM](https://huggingface.co/datasets/codefuse-ai/F2LLM)) covering a diverse range of retrieval, classification, and clustering data, curated solely from open-source datasets without any synthetic data. These models are trained with homogeneous macro batches in a single stage, without sophisticated multi-stage pipelines.
 
 ## Usage