Add library_name, pipeline_tag, and arxiv metadata

This PR improves the model card for [WideSeek-R1: Exploring Width Scaling for Broad Information Seeking via Multi-Agent Reinforcement Learning](https://huggingface.co/papers/2602.04634).

Changes:
- Added `library_name: transformers` to enable proper library detection
- Added `pipeline_tag: text-generation` to enable task filtering and widget support
- Added `arxiv:2602.04634` tag to link the model to the paper page

These metadata improvements will:
1. Make the model discoverable when users filter by text-generation task
2. Link the model to the paper page on Hugging Face
3. Enable the proper inference widget on the model page
4. Help users understand the model's intended use case

Please review and merge if everything looks good.

README.md

@@ -4,6 +4,10 @@ language:
   - en
 base_model:
   - Qwen/Qwen3-4B
+library_name: transformers
+pipeline_tag: text-generation
+tags:
+  - arxiv:2602.04634
 metrics:
   - accuracy
 model-index:
@@ -31,9 +35,9 @@ model-index:
 
 ![image](fig/scaling.png)
 
-Recent advancements in Large Language Models (LLMs) have largely focused on depth scaling, where a single agent solves long-horizon problems with multi-turn reasoning and tool use. However, as tasks grow broader, the key bottleneck shifts from individual competence to organizational capability. 
+Recent advancements in Large Language Models (LLMs) have largely focused on depth scaling, where a single agent solves long-horizon problems with multi-turn reasoning and tool use. However, as tasks grow broader, the key bottleneck shifts from individual competence to organizational capability.
 
-In this work, we explore a complementary dimension of width scaling with multi-agent systems to address broad information seeking. Existing multi-agent systems often rely on hand-crafted workflows and turn-taking interactions that fail to parallelize work effectively. To bridge this gap, we propose WideSeek-R1, a lead-agent-subagent framework trained via multi-agent reinforcement learning (MARL) to synergize scalable orchestration and parallel execution. By utilizing a shared LLM with isolated contexts and specialized tools, WideSeek-R1 jointly optimizes the lead agent and parallel subagents on a curated dataset of 20k broad information-seeking tasks. 
+In this work, we explore a complementary dimension of width scaling with multi-agent systems to address broad information seeking. Existing multi-agent systems often rely on hand-crafted workflows and turn-taking interactions that fail to parallelize work effectively. To bridge this gap, we propose WideSeek-R1, a lead-agent-subagent framework trained via multi-agent reinforcement learning (MARL) to synergize scalable orchestration and parallel execution. By utilizing a shared LLM with isolated contexts and specialized tools, WideSeek-R1 jointly optimizes the lead agent and parallel subagents on a curated dataset of 20k broad information-seeking tasks.
 
 Extensive experiments show that WideSeek-R1-4B achieves an item F1 score of 40.0\% on the WideSearch benchmark, which is comparable to the performance of single-agent DeepSeek-R1-671B. Furthermore, WideSeek-R1-4B exhibits consistent performance gains as the number of parallel subagents increases, highlighting the effectiveness of width scaling.
 
@@ -50,4 +54,4 @@ If you use this model in your research, please cite our paper:
   journal = {arXiv preprint arXiv:2602.04634},
   year    = {2026},
 }
-```
+```