feat: update app

app.py

@@ -57,7 +57,8 @@ i18n = gr.I18n(
         "header": "## 🎤 MOSS Transcribe Diarize: Accurate Transcription with Speaker Diarization",
         "tips": (
             f"> **💡 Note**: This demo currently supports ASR with speaker recognition for audio clips up to **{MAX_AUDIO_DURATION}s**.  \n"
-            "> Stay tuned! Full-featured long audio support (including timestamps) is **Coming Soon**."
+            "> Stay tuned! Full-featured long audio support (including timestamps) is **Coming Soon**.  \n"
+            "> **🔗 Links**: [paper](https://arxiv.org/abs/2601.01554) · [model page](https://mosi.cn/models/moss-transcribe-diarize)"
         ),
         "audio_tab": "🎵 Audio",
         "audio_label": "📥 Upload Audio",
@@ -71,7 +72,8 @@ i18n = gr.I18n(
         "header": "## 🎤 MOSS Transcribe Diarize: 精准转写与说话人识别",
         "tips": (
             f"> **💡 说明**：本演示版本仅支持短音频（最长 **{MAX_AUDIO_DURATION}s**）的文本转写及说话人识别。  \n"
-            "> 敬请期待！支持完整时间戳的长音频 API **即将上线**。"
+            "> 敬请期待！支持完整时间戳的长音频 API **即将上线**。  \n"
+            "> **🔗 链接**：[paper](https://arxiv.org/abs/2601.01554) · [model page](https://mosi.cn/models/moss-transcribe-diarize)"
         ),
         "audio_tab": "🎵 音频",
         "audio_label": "📥 上传音频",