Review

[논문리뷰] Thinking with Drafting: Optical Decompression via Logical Reconstruction

arXiv에 게시된 'Thinking with Drafting: Optical Decompression via Logical Reconstruction' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Reasoning #Visual Algebra #Domain-Specific Language #Optical Decompression #Logical Reconstruction #Bar Model #MLLMs #Verification

2026년 2월 12일

[논문리뷰] ThinkRouter: Efficient Reasoning via Routing Thinking between Latent and Discrete Spaces

Julian McAuley이 arXiv에 게시한 'ThinkRouter: Efficient Reasoning via Routing Thinking between Latent and Discrete Spaces' 논문에 대한 자세한 리뷰입니다.

#Review #Efficient Reasoning #Latent Space Reasoning #Discrete Space Reasoning #LLM Confidence #Routing Mechanism #Inference-Time Optimization #Chain-of-Thought

2026년 2월 12일

[논문리뷰] Think Longer to Explore Deeper: Learn to Explore In-Context via Length-Incentivized Reinforcement Learning

arXiv에 게시된 'Think Longer to Explore Deeper: Learn to Explore In-Context via Length-Incentivized Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #Large Language Models #In-Context Learning #Reinforcement Learning #Test-Time Scaling #Exploration-Exploitation #State Coverage #Reward Shaping #Chain-of-Thought

2026년 2월 12일

[논문리뷰] The Devil Behind Moltbook: Anthropic Safety is Always Vanishing in Self-Evolving AI Societies

Jinyu Hou이 arXiv에 게시한 'The Devil Behind Moltbook: Anthropic Safety is Always Vanishing in Self-Evolving AI Societies' 논문에 대한 자세한 리뷰입니다.

#Review #Multi-agent Systems #Self-evolution #AI Safety #Alignment Drift #Information Theory #Thermodynamics #Entropy Accumulation #Moltbook

2026년 2월 12일

[논문리뷰] Stroke of Surprise: Progressive Semantic Illusions in Vector Sketching

arXiv에 게시된 'Stroke of Surprise: Progressive Semantic Illusions in Vector Sketching' 논문에 대한 자세한 리뷰입니다.

#Review #Vector Sketching #Progressive Semantic Illusions #Score Distillation Sampling #Joint Optimization #Visual Anagrams #Bézier Strokes #CLIP-guided Generation #Diffusion Models

2026년 2월 12일

[논문리뷰] Sparse Video Generation Propels Real-World Beyond-the-View Vision-Language Navigation

Yukuan Xu이 arXiv에 게시한 'Sparse Video Generation Propels Real-World Beyond-the-View Vision-Language Navigation' 논문에 대한 자세한 리뷰입니다.

#Review #Vision-Language Navigation #Beyond-the-View Navigation #Video Generation Models #Sparse Video Generation #Diffusion Models #Embodied AI #Real-world Navigation #Long-horizon Planning

2026년 2월 12일

[논문리뷰] Sci-CoE: Co-evolving Scientific Reasoning LLMs via Geometric Consensus with Sparse Supervision

arXiv에 게시된 'Sci-CoE: Co-evolving Scientific Reasoning LLMs via Geometric Consensus with Sparse Supervision' 논문에 대한 자세한 리뷰입니다.

#Review #LLM #Scientific Reasoning #Co-evolution #Reinforcement Learning #Sparse Supervision #Geometric Consensus #Self-Play #Verifier

2026년 2월 12일

[논문리뷰] ScalSelect: Scalable Training-Free Multimodal Data Selection for Efficient Visual Instruction Tuning

arXiv에 게시된 'ScalSelect: Scalable Training-Free Multimodal Data Selection for Efficient Visual Instruction Tuning' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Data Selection #Visual Instruction Tuning #Training-Free #Scalability #Subspace Learning #Vision-Language Models #Attention Mechanism

2026년 2월 12일

[논문리뷰] RISE: Self-Improving Robot Policy with Compositional World Model

arXiv에 게시된 'RISE: Self-Improving Robot Policy with Compositional World Model' 논문에 대한 자세한 리뷰입니다.

#Review #Robot Learning #Reinforcement Learning #World Models #Compositional Models #Robotic Manipulation #Self-Improving #Vision-Language-Action (VLA)

2026년 2월 12일

[논문리뷰] Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm

arXiv에 게시된 'Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm' 논문에 대한 자세한 리뷰입니다.

#Review #Decentralized Training #Mixture-of-Experts (MoE)#Large Language Models (LLMs)#Memory Efficiency #Sparse Expert Synchronization #Federated Learning #Distributed GPUs

2026년 2월 12일

[논문리뷰] NarraScore: Bridging Visual Narrative and Musical Dynamics via Hierarchical Affective Control

arXiv에 게시된 'NarraScore: Bridging Visual Narrative and Musical Dynamics via Hierarchical Affective Control' 논문에 대한 자세한 리뷰입니다.

#Review #Video-to-Music Generation #Affective Computing #Vision-Language Models (VLMs)#Hierarchical Control #Soundtrack Generation #Temporal Coherence #Emotion-Driven Music

2026년 2월 12일

[논문리뷰] MetaphorStar: Image Metaphor Understanding and Reasoning with End-to-End Visual Reinforcement Learning

Hongsheng Li이 arXiv에 게시한 'MetaphorStar: Image Metaphor Understanding and Reasoning with End-to-End Visual Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #Image Metaphor Understanding #Visual Reasoning #Reinforcement Learning #MLLMs #TFQ-GRPO #End-to-End Learning #Cognitive AI

2026년 2월 12일

[논문리뷰] MOSS-Audio-Tokenizer: Scaling Audio Tokenizers for Future Audio Foundation Models

arXiv에 게시된 'MOSS-Audio-Tokenizer: Scaling Audio Tokenizers for Future Audio Foundation Models' 논문에 대한 자세한 리뷰입니다.

#Review #Audio Tokenizer #Transformer Architecture #End-to-End Learning #Residual Vector Quantization #Speech Synthesis #Audio Foundation Models #Scalability #Autoregressive Models

2026년 2월 12일

[논문리뷰] Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation

arXiv에 게시된 'Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation' 논문에 대한 자세한 리뷰입니다.

#Review #On-Policy Distillation #Reward Extrapolation #Large Language Models (LLMs)#Knowledge Distillation #Reinforcement Learning #Math Reasoning #Code Generation #Multi-teacher Distillation

2026년 2월 12일

[논문리뷰] LawThinker: A Deep Research Legal Agent in Dynamic Environments

arXiv에 게시된 'LawThinker: A Deep Research Legal Agent in Dynamic Environments' 논문에 대한 자세한 리뷰입니다.

#Review #Legal Reasoning #AI Agent #Large Language Models #Verification #Knowledge Management #Dynamic Environments #Procedural Compliance #Tool Use

2026년 2월 12일

[논문리뷰] GigaBrain-0.5M*: a VLA That Learns From World Model-Based Reinforcement Learning

arXiv에 게시된 'GigaBrain-0.5M*: a VLA That Learns From World Model-Based Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #VLA Models #World Models #Reinforcement Learning #Robotic Manipulation #Long-Horizon Control #Human-in-the-Loop #Continual Learning

2026년 2월 12일

[논문리뷰] EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration

Yinghui Li이 arXiv에 게시한 'EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration' 논문에 대한 자세한 리뷰입니다.

#Review #Humanoid Robotics #Loco-Manipulation #Egocentric Demonstration #Robot-Free Learning #Cross-Embodiment Transfer #View Alignment #Action Alignment #VLA Co-training

2026년 2월 12일

[논문리뷰] DeepSight: An All-in-One LM Safety Toolkit

arXiv에 게시된 'DeepSight: An All-in-One LM Safety Toolkit' 논문에 대한 자세한 리뷰입니다.

#Review #LM Safety #Evaluation #Diagnosis #Multimodal AI #Frontier AI Risks #Black-box Analysis #White-box Insight #Open-source Toolkit

2026년 2월 12일

[논문리뷰] DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing

arXiv에 게시된 'DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Model #Image Generation #Image Editing #Diffusion Models #VLM-DiT Architecture #Stacked Channel Bridging #Reinforcement Learning #Lightweight Models

2026년 2월 12일

[논문리뷰] Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models

arXiv에 게시된 'Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #Large Language Models #Prompt Engineering #Compositional Generalization #Verifiable Rewards #Curriculum Learning #Mathematical Reasoning #Multi-task Learning

2026년 2월 12일