#Model Efficiency

12개의 포스트

[논문리뷰] Mobile-O: Unified Multimodal Understanding and Generation on Mobile Device

arXiv에 게시된 'Mobile-O: Unified Multimodal Understanding and Generation on Mobile Device' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal AI #Vision-Language Models #Diffusion Models #Mobile Devices #Edge Computing #Model Efficiency #Unified Architecture #Real-time Inference

2026년 2월 23일

[논문리뷰] 2Mamba2Furious: Linear in Complexity, Competitive in Accuracy

Eric C. Larson이 arXiv에 게시한 '2Mamba2Furious: Linear in Complexity, Competitive in Accuracy' 논문에 대한 자세한 리뷰입니다.

#Review #Linear Attention #Mamba-2 #High-Order Attention #Model Efficiency #Long Context #Transformer #State Space Models

2026년 2월 19일

[논문리뷰] Routing the Lottery: Adaptive Subnetworks for Heterogeneous Data

Michal Byra이 arXiv에 게시한 'Routing the Lottery: Adaptive Subnetworks for Heterogeneous Data' 논문에 대한 자세한 리뷰입니다.

#Review #Pruning #Lottery Ticket Hypothesis #Adaptive Subnetworks #Heterogeneous Data #Model Efficiency #Conditional Computation #Subnetwork Collapse

2026년 2월 1일

[논문리뷰] VIBE: Visual Instruction Based Editor

Bulat Suleimanov이 arXiv에 게시한 'VIBE: Visual Instruction Based Editor' 논문에 대한 자세한 리뷰입니다.

#Review #Instruction-Based Image Editing #Diffusion Models #Vision-Language Models (VLM)#Model Efficiency #Multi-stage Training #Preference Alignment #Source Consistency

2026년 1월 15일

[논문리뷰] STEP3-VL-10B Technical Report

arXiv에 게시된 'STEP3-VL-10B Technical Report' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Large Language Models #Vision-Language Models #Reinforcement Learning #Parallel Coordinated Reasoning #Model Efficiency #Foundation Models #Pre-training #Post-training

2026년 1월 15일

[논문리뷰] Universal Reasoning Model

arXiv에 게시된 'Universal Reasoning Model' 논문에 대한 자세한 리뷰입니다.

#Review #Universal Transformer #Recurrent Neural Networks #ARC-AGI #Reasoning Tasks #Nonlinearity #Convolutional Gating #Truncated Backpropagation #Model Efficiency

2025년 12월 17일

[논문리뷰] SUCCESS-GS: Survey of Compactness and Compression for Efficient Static and Dynamic Gaussian Splatting

Sung-Ho Bae이 arXiv에 게시한 'SUCCESS-GS: Survey of Compactness and Compression for Efficient Static and Dynamic Gaussian Splatting' 논문에 대한 자세한 리뷰입니다.

#Review #3D Gaussian Splatting (3DGS)#Gaussian Compression #Model Efficiency #Novel View Synthesis #Dynamic Scenes #Parameter Compression #Restructuring Compression #Real-time Rendering

2025년 12월 9일

[논문리뷰] Script: Graph-Structured and Query-Conditioned Semantic Token Pruning for Multimodal Large Language Models

arXiv에 게시된 'Script: Graph-Structured and Query-Conditioned Semantic Token Pruning for Multimodal Large Language Models' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Large Language Models (MLLMs)#Token Pruning #Graph-Structured Pruning (GSP)#Query-Conditioned Semantic Pruning (QCSP)#Determinantal Point Processes (DPP)#Model Efficiency #Visual Redundancy

2025년 12월 1일

[논문리뷰] Flash-DMD: Towards High-Fidelity Few-Step Image Generation with Efficient Distillation and Joint Reinforcement Learning

arXiv에 게시된 'Flash-DMD: Towards High-Fidelity Few-Step Image Generation with Efficient Distillation and Joint Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion Models #Image Generation #Distillation #Reinforcement Learning #Few-Step Sampling #Timestep-Aware #Pixel-GAN #Model Efficiency

2025년 12월 1일

[논문리뷰] Virtual Width Networks

arXiv에 게시된 'Virtual Width Networks' 논문에 대한 자세한 리뷰입니다.

#Review #Virtual Width Networks #Transformer #Mixture-of-Experts (MoE)#Scaling Laws #Representation Learning #Model Efficiency #Multi-Token Prediction #Hyper-Connections

2025년 11월 16일

[논문리뷰] Shorter but not Worse: Frugal Reasoning via Easy Samples as Length Regularizers in Math RLVR

arXiv에 게시된 'Shorter but not Worse: Frugal Reasoning via Easy Samples as Length Regularizers in Math RLVR' 논문에 대한 자세한 리뷰입니다.

#Review #LLMs #RLVR #Length Regularization #Mathematical Reasoning #Data Curation #Model Efficiency #Emergent Brevity

2025년 11월 9일

[논문리뷰] Are We Using the Right Benchmark: An Evaluation Framework for Visual Token Compression Methods

Yiyu Wang이 arXiv에 게시한 'Are We Using the Right Benchmark: An Evaluation Framework for Visual Token Compression Methods' 논문에 대한 자세한 리뷰입니다.

#Review #Visual Token Compression #MLLMs #Evaluation Framework #Benchmarking #Downsampling #Data Filtering #Model Efficiency

2025년 10월 9일