#Self-Attention

8개의 포스트

[논문리뷰] VISion On Request: Enhanced VLLM efficiency with sparse, dynamically selected, vision-language interactions

Georgios Tzimiropoulos이 arXiv에 게시한 'VISion On Request: Enhanced VLLM efficiency with sparse, dynamically selected, vision-language interactions' 논문에 대한 자세한 리뷰입니다.

#Review #LVLM Efficiency #Sparse Interaction #Cross-Attention #Self-Attention #Adaptive Inference #Visual Feature Refinement #Computational Cost Reduction

2026년 3월 24일

[논문리뷰] Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs

arXiv에 게시된 'Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Rotary Position Embedding #Long-Context LLMs #Complex-Valued Neural Networks #Self-Attention #Positional Encoding #Information Loss #Length Extrapolation

2025년 12월 8일

[논문리뷰] Toward the Frontiers of Reliable Diffusion Sampling via Adversarial Sinkhorn Attention Guidance

Kwanyoung Kim이 arXiv에 게시한 'Toward the Frontiers of Reliable Diffusion Sampling via Adversarial Sinkhorn Attention Guidance' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion Models #Guidance Sampling #Optimal Transport #Sinkhorn Algorithm #Self-Attention #Adversarial Perturbation #Image Generation #ControlNet

2025년 11월 12일

[논문리뷰] ACG: Action Coherence Guidance for Flow-based VLA models

arXiv에 게시된 'ACG: Action Coherence Guidance for Flow-based VLA models' 논문에 대한 자세한 리뷰입니다.

#Review #Action Coherence #Flow Matching #VLA Models #Guidance #Robotics #Imitation Learning #Transformer #Self-Attention

2025년 10월 28일

[논문리뷰] Sparser Block-Sparse Attention via Token Permutation

arXiv에 게시된 'Sparser Block-Sparse Attention via Token Permutation' 논문에 대한 자세한 리뷰입니다.

#Review #Large Language Models (LLMs)#Self-Attention #Block-Sparse Attention #Token Permutation #Computational Efficiency #Prefilling #Long Context #Causal Attention

2025년 10월 27일

[논문리뷰] HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives

arXiv에 게시된 'HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives' 논문에 대한 자세한 리뷰입니다.

#Review #Text-to-Video Generation #Multi-Shot Video #Narrative Coherence #Diffusion Models #Self-Attention #Cinematic AI #Video Consistency #Directorial Control

2025년 10월 24일

[논문리뷰] DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models

Roi Reichart이 arXiv에 게시한 'DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models' 논문에 대한 자세한 리뷰입니다.

#Review #Semantic Leakage #Text-to-Image Models #Attention Control #Inference-time Mitigation #Diffusion Models #Evaluation Dataset #Self-Attention

2025년 10월 23일

[논문리뷰] Stand-In: A Lightweight and Plug-and-Play Identity Control for Video Generation

Chen Li이 arXiv에 게시한 'Stand-In: A Lightweight and Plug-and-Play Identity Control for Video Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Video Generation #Identity Preservation #Plug-and-Play #Diffusion Models #Self-Attention #Lightweight AI #Conditional Image Branch

2025년 8월 14일