作者位次 #4/7ICLR 2026Poster· Applications / Language, Speech and Dialog
MARS-Sep: Multimodal-Aligned Reinforced Sound Separation
Zihan Zhang · Xize Cheng · Zhennan Jiang · Dongjie Fu · Jingyuan Chen · Zhou Zhao · Tao Jin
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Zihan Zhang · Xize Cheng · Zhennan Jiang · Dongjie Fu · Jingyuan Chen · Zhou Zhao · Tao Jin
Fangming Feng · Dongjie Fu · Zequn Xie · Yu Zhang · Yangyang Wu · Zhou Zhao · Tao Jin
While diffusion and flow-matching models have advanced TTS, generating high-arousal emotions remains a persistent challenge due to the trade-off between stability and expressiveness. Existing systems often suffer from linguistic collapse when pursuing high intensity or fail to meet target emotion...