Add paper/code links, update task category, and add citation

README.md

@@ -1,33 +1,33 @@
 ---
 license: apache-2.0
-pretty_name: SyncReward-Data-80K
+size_categories:
+- 10K<n<100K
 task_categories:
-  - video-classification
+- other
+pretty_name: SyncReward-Data-80K
 tags:
-  - audio-video
-  - audio-visual
-  - synchronization
-  - reward-model
-  - human-feedback
-  - video-generation
-size_categories:
-  - 10K<n<100K
+- audio-video
+- audio-visual
+- synchronization
+- reward-model
+- human-feedback
+- video-generation
 configs:
-  - config_name: annotations
-    default: true
-    data_files:
-      - split: train
-        path: annotations/annotations_80k.jsonl
-  - config_name: stage2
-    data_files:
-      - split: train
-        path: splits/stage2_train.jsonl
-      - split: validation
-        path: splits/stage2_val.jsonl
-  - config_name: bench
-    data_files:
-      - split: test
-        path: bench/syncreward_bench.jsonl
+- config_name: annotations
+  default: true
+  data_files:
+  - split: train
+    path: annotations/annotations_80k.jsonl
+- config_name: stage2
+  data_files:
+  - split: train
+    path: splits/stage2_train.jsonl
+  - split: validation
+    path: splits/stage2_val.jsonl
+- config_name: bench
+  data_files:
+  - split: test
+    path: bench/syncreward_bench.jsonl
 ---
 
 # SyncReward-Data-80K
@@ -41,7 +41,8 @@ This dataset contains:
 
 It is the data behind the [SyncReward](https://huggingface.co/Guan123/SyncReward) reward model.
 
-Paper, code repository and citation: coming soon.
+**Paper:** [SyncReward: Human-Aligned Synchronization Reward for Audio-Visual Generation](https://huggingface.co/papers/2610.16338)  
+**Code:** [https://github.com/WeChatCV/SyncReward](https://github.com/WeChatCV/SyncReward)
 
 ## Repository layout
 
@@ -191,4 +192,16 @@ generation evaluation.
 
 ## Citation
 
-Coming soon.
+Please cite our paper if you use this dataset:
+
+```bibtex
+@misc{guan2026syncrewardhumanalignedsynchronizationreward,
+  title={SyncReward: Human-Aligned Synchronization Reward for Audio-Visual Generation},
+  author={Kaisi Guan and Ying Ba and Wenjing Wang and Xihua Wang and Chen Li and Jing Lyu and Ruihau Song},
+  year={2026},
+  eprint={2610.16338},
+  archivePrefix={arXiv},
+  primaryClass={cs.CV},
+  url={https://arxiv.org/abs/2610.16338},
+}
+```