Add library_name metadata and improve model card structure

This PR adds the `library_name: transformers` to the metadata section. This enables the "Use in Transformers" button and automated code snippets on the model page, making it easier for users to load the model.

The model card has also been updated to clearly reference the paper [AWED-FiNER: Agents, Web applications, and Expert Detectors for Fine-grained Named Entity Recognition across 36 Languages for 6.6 Billion Speakers](https://huggingface.co/papers/2601.10161), as this model is one of the expert detectors developed in that work.

README.md

@@ -1,16 +1,17 @@
 ---
-license: mit
+base_model:
+- FacebookAI/xlm-roberta-large
 datasets:
 - MultiCoNER/multiconer_v2
 language:
 - zh
+license: mit
 metrics:
 - f1
 - precision
 - recall
-base_model:
-- FacebookAI/xlm-roberta-large
 pipeline_tag: token-classification
+library_name: transformers
 tags:
 - NER
 - Named_Entity_Recognition
@@ -19,6 +20,8 @@ pretty_name: MultiCoNER2 Chinese XLM-RoBERTa
 
 **XLM-RoBERTa is fine-tuned on Chinese [MultiCoNER2](https://huggingface.co/datasets/MultiCoNER/multiconer_v2) dataset for Fine-grained Named Entity Recognition.**
 
+This model is an expert detector part of the **AWED-FiNER** project, as described in the paper: [AWED-FiNER: Agents, Web applications, and Expert Detectors for Fine-grained Named Entity Recognition across 36 Languages for 6.6 Billion Speakers](https://huggingface.co/papers/2601.10161).
+
 The tagset of [MultiCoNER2](https://huggingface.co/datasets/MultiCoNER/multiconer_v2) is a fine-grained tagset. The fine to coarse level mapping of the tags are as follows:
 
   * Location (LOC) : Facility, OtherLOC, HumanSettlement, Station
@@ -93,4 +96,5 @@ If you use this model, please cite the following papers:
   booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
   volume={40},
   year={2026}
-}
+}
+```