#End-to-End

5개의 포스트

[논문리뷰] WavAlign: Enhancing Intelligence and Expressiveness in Spoken Dialogue Models via Adaptive Hybrid Post-Training

Yangzhuo Li이 arXiv에 게시한 'WavAlign: Enhancing Intelligence and Expressiveness in Spoken Dialogue Models via Adaptive Hybrid Post-Training' 논문에 대한 자세한 리뷰입니다.

#Review #Spoken Dialogue Models #Post-Training #Reinforcement Learning #Preference Optimization #Modality Alignment #End-to-End #Acoustic Expressiveness

2026년 4월 22일

[논문리뷰] AURA: Always-On Understanding and Real-Time Assistance via Video Streams

Xintong Guo이 arXiv에 게시한 'AURA: Always-On Understanding and Real-Time Assistance via Video Streams' 논문에 대한 자세한 리뷰입니다.

#Review #VideoLLMs #Streaming Video Understanding #End-to-End #Context Management #Proactive Response #Real-Time Inference

2026년 4월 6일

[논문리뷰] MonoArt: Progressive Structural Reasoning for Monocular Articulated 3D Reconstruction

arXiv에 게시된 'MonoArt: Progressive Structural Reasoning for Monocular Articulated 3D Reconstruction' 논문에 대한 자세한 리뷰입니다.

#Review #Monocular 3D Reconstruction #Articulated Objects #Progressive Structural Reasoning #Kinematic Estimation #PartNet-Mobility #End-to-End

2026년 3월 19일

[논문리뷰] ViSAudio: End-to-End Video-Driven Binaural Spatial Audio Generation

arXiv에 게시된 'ViSAudio: End-to-End Video-Driven Binaural Spatial Audio Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Binaural Audio Generation #Spatial Audio #Video-Driven #End-to-End #Conditional Flow Matching #Multimodal AI #Deep Learning #Audio-Visual Synthesis

2025년 12월 2일

[논문리뷰] OmniFusion: Simultaneous Multilingual Multimodal Translations via Modular Fusion

arXiv에 게시된 'OmniFusion: Simultaneous Multilingual Multimodal Translations via Modular Fusion' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Translation #Speech Translation #Simultaneous Translation #Large Language Models #Multimodal Foundation Models #Modular Fusion #End-to-End #Gated Fusion #OCR

2025년 12월 1일