Add library_name and base_model metadata

#1
by nielsr HF Staff - opened
Files changed (1) hide show
  1. README.md +8 -2
README.md CHANGED
@@ -1,17 +1,21 @@
1
  ---
2
- license: apache-2.0
3
  datasets:
4
  - Zichen1024/CoVe-12k
 
5
  pipeline_tag: text-generation
 
 
6
  tags:
7
  - agent
 
8
  ---
 
9
  <p align="center">
10
  <img src="https://cdn-uploads.huggingface.co/production/uploads/670a8557222579c05ec3005c/Mxt8wCxBLOs674WQ0jMIG.png" alt="CoVe Mascot" width="400"/>
11
  </p>
12
 
13
  <p align="center">
14
- <a href="https://arxiv.org/abs/2603.01940">πŸ“„ Paper</a> &nbsp;|&nbsp;
15
  <a href="https://cove-agent.github.io">🌐 Website</a> &nbsp;|&nbsp;
16
  <a href="https://huggingface.co/datasets/Zichen1024/CoVe-12k">πŸ€— Dataset</a> &nbsp;|&nbsp;
17
  </p>
@@ -20,6 +24,8 @@ tags:
20
 
21
  **CoVe-4B** is a compact 4B interactive tool-use agent fine-tuned from [Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507) using the **CoVe** (Constraint-Verification) post-training framework. It is trained on [CoVe-12K](https://huggingface.co/datasets/Zichen1024/CoVe-12k), a dataset of 12K high-quality multi-turn tool-use trajectories synthesized and verified by deterministic constraint checking.
22
 
 
 
23
  ## Framework
24
 
25
  <p style="text-align: justify;">
 
1
  ---
 
2
  datasets:
3
  - Zichen1024/CoVe-12k
4
+ license: apache-2.0
5
  pipeline_tag: text-generation
6
+ library_name: transformers
7
+ base_model: Qwen/Qwen3-4B-Instruct-2507
8
  tags:
9
  - agent
10
+ - tool-use
11
  ---
12
+
13
  <p align="center">
14
  <img src="https://cdn-uploads.huggingface.co/production/uploads/670a8557222579c05ec3005c/Mxt8wCxBLOs674WQ0jMIG.png" alt="CoVe Mascot" width="400"/>
15
  </p>
16
 
17
  <p align="center">
18
+ <a href="https://huggingface.co/papers/2603.01940">πŸ“„ Paper</a> &nbsp;|&nbsp;
19
  <a href="https://cove-agent.github.io">🌐 Website</a> &nbsp;|&nbsp;
20
  <a href="https://huggingface.co/datasets/Zichen1024/CoVe-12k">πŸ€— Dataset</a> &nbsp;|&nbsp;
21
  </p>
 
24
 
25
  **CoVe-4B** is a compact 4B interactive tool-use agent fine-tuned from [Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507) using the **CoVe** (Constraint-Verification) post-training framework. It is trained on [CoVe-12K](https://huggingface.co/datasets/Zichen1024/CoVe-12k), a dataset of 12K high-quality multi-turn tool-use trajectories synthesized and verified by deterministic constraint checking.
26
 
27
+ The model was introduced in the paper [CoVe: Training Interactive Tool-Use Agents via Constraint-Guided Verification](https://huggingface.co/papers/2603.01940).
28
+
29
  ## Framework
30
 
31
  <p style="text-align: justify;">