Best Papers
The 2023 International Doctoral Forum received a total of 103 papers, with over 100 professors and students participating in the conference. After a thorough peer review process of the submitted papers and oral presentations, 12 outstanding papers were selected as the best papers. The list of winning papers and their authors are as follows:
- Title: MSMC-TTS: Multi-Stage Multi-Codebook VQ-VAE Based Neural TTS
- Speaker: Haohan Guo
- Title: MSStyleTTS: Multi-Scale Style Modeling with Hierarchical Context Information for Expressive Speech Synthesis
- Speaker: Yixuan Zhou
- Title: Disentangling Voice and Content with Self-Supervision for Speaker Recognition
- Speaker: Tianchi Liu
- Title: UniSyn: An End-to-End Unified Model for Text-to-Speech and Singing Voice Synthesis
- Speaker: Yi Lei
- Title: NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation
- Speaker: Shengming Yin
- Title: VaBUS: Edge-Cloud Real-Time Video Analytics via Background Understanding and Subtraction
- Speaker: Hanling Wang
- Title: VI-Map: Infrastructure-Assisted Real-Time HD Mapping for Autonomous Driving
- Speaker: Yuze He
- Title: Fast Lossless Neural Compression with Integer-Only Discrete Flows
- Speaker: Siyu Wang
- Title: Learning from Limited Heterogeneous Training Data: Meta-Learning for Unsupervised Zero-Day Web Attack Detection across Web Domains
- Speaker: Peiyang Li
- Title: Detecting Unknown Encrypted Malicious Traffic in Real Time via Flow Interaction Graph Analysis
- Speaker: Chuanpu Fu
- Title: BaGuaLu: Targeting Brain Scale Pretrained Models with over 37 Million Cores
- Speaker: Zixuan Ma
- Title: Large Language Models as Source Planner for Personalized Knowledge-grounded Dialogues
- Speaker: Hongru Wang