#Efficiency

31개의 포스트

[논문리뷰] Apriel-Reasoner: RL Post-Training for General-Purpose and Efficient Reasoning

arXiv에 게시된 'Apriel-Reasoner: RL Post-Training for General-Purpose and Efficient Reasoning' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #Chain-of-Thought #Efficiency #RLVR #Multi-Domain #Reasoning

2026년 4월 2일

[논문리뷰] AVControl: Efficient Framework for Training Audio-Visual Controls

arXiv에 게시된 'AVControl: Efficient Framework for Training Audio-Visual Controls' 논문에 대한 자세한 리뷰입니다.

#Review #Audio-Visual Generation #Video Control #LoRA #Parallel Canvas Conditioning #Diffusion Models #Modularity #Efficiency

2026년 3월 26일

[논문리뷰] Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs

Ehud Rivlin이 arXiv에 게시한 'Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Vision-language models #High-resolution crops #Tool-calling #Efficiency #Reinforcement Learning #Multi-turn interaction #Data curation #Spatial-on-demand

2026년 3월 23일

[논문리뷰] Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization

arXiv에 게시된 'Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization' 논문에 대한 자세한 리뷰입니다.

#Review #Agentic AI #Long-Horizon Search #Parallel Execution #Data Synthesis #Reinforcement Learning #Generalization #Efficiency #LLM Agent

2026년 2월 26일

[논문리뷰] SimVLA: A Simple VLA Baseline for Robotic Manipulation

arXiv에 게시된 'SimVLA: A Simple VLA Baseline for Robotic Manipulation' 논문에 대한 자세한 리뷰입니다.

#Review #Robotic Manipulation #Vision-Language-Action (VLA) Models #Baseline Model #Modular Design #Flow Matching #Zero-Shot Generalization #Standardized Training #Efficiency

2026년 2월 23일

[논문리뷰] InftyThink+: Effective and Efficient Infinite-Horizon Reasoning via Reinforcement Learning

arXiv에 게시된 'InftyThink+: Effective and Efficient Infinite-Horizon Reasoning via Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #Iterative Reasoning #Reinforcement Learning #Large Language Models #Context Management #Summarization #Chain-of-Thought #Efficiency #Mathematical Reasoning

2026년 2월 8일

[논문리뷰] OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models

Yiyan Ji이 arXiv에 게시한 'OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models' 논문에 대한 자세한 리뷰입니다.

#Review #Omni-modal LLMs #Token Compression #Modality-Asymmetric #Video Pruning #Audio Selection #Efficiency #Large Language Models #Spatio-Temporal

2026년 2월 4일

[논문리뷰] Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection

Jae-Joon Kim이 arXiv에 게시한 'Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection' 논문에 대한 자세한 리뷰입니다.

#Review #Sparse Attention #Long-Context Inference #LLMs #Token Selection #Efficiency #Transformer #Dynamic Sparsity

2026년 2월 3일

[논문리뷰] ConceptMoE: Adaptive Token-to-Concept Compression for Implicit Compute Allocation

arXiv에 게시된 'ConceptMoE: Adaptive Token-to-Concept Compression for Implicit Compute Allocation' 논문에 대한 자세한 리뷰입니다.

#Review #MoE #LLMs #Adaptive Compression #Token Merging #Compute Allocation #Efficiency #Vision-Language Models #Continual Training

2026년 1월 29일

[논문리뷰] UPLiFT: Efficient Pixel-Dense Feature Upsampling with Local Attenders

arXiv에 게시된 'UPLiFT: Efficient Pixel-Dense Feature Upsampling with Local Attenders' 논문에 대한 자세한 리뷰입니다.

#Review #Feature Upsampling #Local Attender #Pixel-Dense Features #Iterative Upsampling #Vision Transformer #Efficiency #Generative AI #Semantic Segmentation

2026년 1월 28일

[논문리뷰] Transition Matching Distillation for Fast Video Generation

arXiv에 게시된 'Transition Matching Distillation for Fast Video Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Video Generation #Diffusion Models #Model Distillation #Few-Step Sampling #Transition Matching #Flow Matching #DMD2 #Efficiency

2026년 1월 15일

[논문리뷰] HiStream: Efficient High-Resolution Video Generation via Redundancy-Eliminated Streaming

arXiv에 게시된 'HiStream: Efficient High-Resolution Video Generation via Redundancy-Eliminated Streaming' 논문에 대한 자세한 리뷰입니다.

#Review #High-Resolution Video Generation #Diffusion Models #Autoregressive #Efficiency #Caching #Attention Mechanisms #Video Streaming #Temporal Consistency

2025년 12월 24일

[논문리뷰] InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models

arXiv에 게시된 'InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models' 논문에 대한 자세한 리뷰입니다.

#Review #Vision-Language Models #Linear Attention #Sliding Window Attention #Gated DeltaNet #Long-Context Understanding #Efficiency #Hybrid Architecture #Multimodal Learning

2025년 12월 10일

[논문리뷰] LongCat-Image Technical Report

arXiv에 게시된 'LongCat-Image Technical Report' 논문에 대한 자세한 리뷰입니다.

#Review #Image Generation #Text-to-Image #Image Editing #Diffusion Model #Multilingual Text Rendering #Photorealism #Efficiency #Open-Source

2025년 12월 8일

[논문리뷰] Focused Chain-of-Thought: Efficient LLM Reasoning via Structured Input Information

Kristian Kersting이 arXiv에 게시한 'Focused Chain-of-Thought: Efficient LLM Reasoning via Structured Input Information' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Reasoning #Chain-of-Thought #Prompt Engineering #Efficiency #Structured Input #Information Extraction #Cognitive Psychology #Token Reduction

2025년 11월 30일

[논문리뷰] Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal Models

Shijie Dong이 arXiv에 게시한 'Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal Models' 논문에 대한 자세한 리뷰입니다.

#Review #Large Multimodal Models #Visual Token Compression #Token Pruning #Benchmark #Efficiency #Inference Latency #Multimodal LLMs

2025년 11월 9일

[논문리뷰] PixelRefer: A Unified Framework for Spatio-Temporal Object Referring with Arbitrary Granularity

Kehan Li이 arXiv에 게시한 'PixelRefer: A Unified Framework for Spatio-Temporal Object Referring with Arbitrary Granularity' 논문에 대한 자세한 리뷰입니다.

#Review #MLLM #Region-level Understanding #Object-centric Reasoning #Spatio-temporal Referring #Video Understanding #Scale-Adaptive Tokenizer #Efficiency #Instruction Tuning

2025년 10월 28일

[논문리뷰] E^2Rank: Your Text Embedding can Also be an Effective and Efficient Listwise Reranker

arXiv에 게시된 'E^2Rank: Your Text Embedding can Also be an Effective and Efficient Listwise Reranker' 논문에 대한 자세한 리뷰입니다.

#Review #Text Embedding #Listwise Reranking #Information Retrieval #Pseudo Relevance Feedback #Contrastive Learning #Multi-task Learning #Efficiency #LLM-based Ranking

2025년 10월 28일

[논문리뷰] LaSeR: Reinforcement Learning with Last-Token Self-Rewarding

arXiv에 게시된 'LaSeR: Reinforcement Learning with Last-Token Self-Rewarding' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #LLM #Self-Verification #Last-Token #Reward Modeling #Efficiency #Reasoning #RLVR

2025년 10월 17일

[논문리뷰] Native Hybrid Attention for Efficient Sequence Modeling

Yu Cheng이 arXiv에 게시한 'Native Hybrid Attention for Efficient Sequence Modeling' 논문에 대한 자세한 리뷰입니다.

#Review #Sequence Modeling #Hybrid Attention #Transformer Architecture #Linear Attention #Sliding Window Attention #Long Context #Large Language Models (LLMs)#Efficiency

2025년 10월 9일

[논문리뷰] MixReasoning: Switching Modes to Think

arXiv에 게시된 'MixReasoning: Switching Modes to Think' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Reasoning #Chain-of-Thought #Efficiency #LoRA #Adaptive Reasoning #Token Uncertainty #Dynamic Switching #Reasoning Compression

2025년 10월 8일

[논문리뷰] Efficient Multi-modal Large Language Models via Progressive Consistency Distillation

arXiv에 게시된 'Efficient Multi-modal Large Language Models via Progressive Consistency Distillation' 논문에 대한 자세한 리뷰입니다.

#Review #Multi-modal LLMs #Token Compression #Efficiency #Knowledge Distillation #Progressive Learning #Consistency Distillation #MLLM Training

2025년 10월 6일

[논문리뷰] Apriel-1.5-15b-Thinker

arXiv에 게시된 'Apriel-1.5-15b-Thinker' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Reasoning Model #Open-Weights Model #Continual Pretraining (CPT)#Supervised Fine-Tuning (SFT)#Training Design #Efficiency #Frontier Performance

2025년 10월 6일

[논문리뷰] GUI-KV: Efficient GUI Agents via KV Cache with Spatio-Temporal Awareness

Chien-Sheng Wu이 arXiv에 게시한 'GUI-KV: Efficient GUI Agents via KV Cache with Spatio-Temporal Awareness' 논문에 대한 자세한 리뷰입니다.

#Review #GUI Agents #KV Cache Compression #Spatio-Temporal Awareness #Vision-Language Models #Efficiency #Attention Sparsity #QR Decomposition

2025년 10월 2일

[논문리뷰] Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel Execution

arXiv에 게시된 'Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel Execution' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Agents #Parallel Execution #DAG-based Planning #Tool Orchestration #Web Agents #Reasoning Framework #Efficiency

2025년 10월 2일

[논문리뷰] Efficient Audio-Visual Speech Separation with Discrete Lip Semantics and Multi-Scale Global-Local Attention

arXiv에 게시된 'Efficient Audio-Visual Speech Separation with Discrete Lip Semantics and Multi-Scale Global-Local Attention' 논문에 대한 자세한 리뷰입니다.

#Review #Audio-Visual Speech Separation #Deep Learning #Efficiency #Discrete Lip Semantics #Global-Local Attention #Lightweight Models #VQ-VAE

2025년 10월 1일

[논문리뷰] MetaEmbed: Scaling Multimodal Retrieval at Test-Time with Flexible Late Interaction

Xintao Chen이 arXiv에 게시한 'MetaEmbed: Scaling Multimodal Retrieval at Test-Time with Flexible Late Interaction' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Retrieval #Late Interaction #Meta Tokens #Matryoshka Representation Learning #Test-Time Scaling #Vision-Language Models #Dense Retrieval #Efficiency

2025년 9월 23일

[논문리뷰] FastFit: Accelerating Multi-Reference Virtual Try-On via Cacheable Diffusion Models

Zhen Wang이 arXiv에 게시한 'FastFit: Accelerating Multi-Reference Virtual Try-On via Cacheable Diffusion Models' 논문에 대한 자세한 리뷰입니다.

#Review #Virtual Try-On #Diffusion Models #Cacheable Architecture #Multi-Reference #Semi-Attention #Efficiency #Image Synthesis

2025년 9월 3일

[논문리뷰] R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning

Han Hu이 arXiv에 게시한 'R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Large Language Models (MLLMs)#Auto-Thinking #Reinforcement Learning (RL)#Bi-mode Annealing #Bi-mode Policy Optimization (BPO)#General-Purpose AI #Reasoning #Efficiency

2025년 9월 1일

[논문리뷰] Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning

Baihong Yuan이 arXiv에 게시한 'Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning' 논문에 대한 자세한 리뷰입니다.

#Review #Sparse Attention #LLMs #Reasoning Tasks #Efficiency #Training-Free #Global Locality #KV Cache Optimization

2025년 8월 12일

[논문리뷰] Pruning the Unsurprising: Efficient Code Reasoning via First-Token Surprisal

Chengcheng Wan이 arXiv에 게시한 'Pruning the Unsurprising: Efficient Code Reasoning via First-Token Surprisal' 논문에 대한 자세한 리뷰입니다.

#Review #Code Reasoning #CoT Compression #LLMs #Efficiency #Surprisal #Pruning #Fine-tuning #Large Reasoning Models

2025년 8월 11일