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

This PR enhances the model card for the `sk2decompile-struct-6.7b` model by:

- Adding the `pipeline_tag: text-generation` to help users discover the model under relevant categories.
- Specifying the `library_name: transformers`, enabling the automated "How to use" code snippet on the Hugging Face Hub, as the model's `config.json` and `tokenizer_config.json` confirm `transformers` compatibility (e.g., `LlamaForCausalLM` architecture).
- Including a direct link to the paper: [Decompile-Bench: Million-Scale Binary-Source Function Pairs for Real-World Binary Decompilation](https://huggingface.co/papers/2505.12668).
- Providing a link to the associated GitHub repository: https://github.com/albertan017/LLM4Decompile.

The existing usage section has been retained as it directly refers to this model and its two-phase operation, and no `transformers`-based usage for the specific "structure recovery" task was found without making modifications to prompts, which is against the guidelines.

README.md

@@ -1,7 +1,14 @@
 ---
 license: mit
+pipeline_tag: text-generation
+library_name: transformers
 ---
-SK²Decompile: LLM-based Two-Phase Binary Decompilation from Skeleton to Skin
+
+# SK²Decompile: Structure Recovery Model
+
+This repository contains the `sk2decompile-struct-6.7b` model, which is part of the SK²Decompile framework introduced in the paper [Decompile-Bench: Million-Scale Binary-Source Function Pairs for Real-World Binary Decompilation](https://huggingface.co/papers/2505.12668).
+
+The full codebase and more details can be found on the [GitHub repository](https://github.com/albertan017/LLM4Decompile).
 
 SK²Decompile is a novel two-phase framework for binary decompilation using Large Language Models (LLMs). Our approach decomposes the complex decompilation task into two manageable phases:
 
@@ -13,4 +20,4 @@ Usage:
 ```
 python normalize_pseudo.py --input_json reverse_sample.json --output_json reverse_sample.json
 python sk2decompile.py --dataset_path reverse_sample.json --model_path LLM4Binary/sk2decompile-struct-6.7b --recover_model_path LLM4Binary/sk2decompile-ident-6.7
-```
+```