Review

[논문리뷰] Project Ariadne: A Structural Causal Framework for Auditing Faithfulness in LLM Agents

arXiv에 게시된 'Project Ariadne: A Structural Causal Framework for Auditing Faithfulness in LLM Agents' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Agents #Faithfulness #XAI #Causal Inference #Structural Causal Models #Counterfactual Interventions #Reasoning Trace Auditing #Causal Decoupling

2026년 1월 5일

[논문리뷰] OpenNovelty: An LLM-powered Agentic System for Verifiable Scholarly Novelty Assessment

Chunchun Ma이 arXiv에 게시한 'OpenNovelty: An LLM-powered Agentic System for Verifiable Scholarly Novelty Assessment' 논문에 대한 자세한 리뷰입니다.

#Review #LLM 에이전트 시스템 #학술 독창성 평가 #피어 리뷰 지원 #증거 기반 검증 #의미론적 검색 #계층적 분류 체계 #대규모 언어 모델

2026년 1월 5일

[논문리뷰] NextFlow: Unified Sequential Modeling Activates Multimodal Understanding and Generation

arXiv에 게시된 'NextFlow: Unified Sequential Modeling Activates Multimodal Understanding and Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal AI #Decoder-only Transformer #Next-scale Prediction #Image Generation #Image Editing #Reinforcement Learning #Unified Modeling #TokenFlow

2026년 1월 5일

[논문리뷰] M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Models

Jun-Cheng Chen이 arXiv에 게시한 'M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Models' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion Models #Concept Erasure #Multimodal Evaluation #Adversarial Attacks #Robustness #Textual Inversion #Latent Inversion #Cross-Attention

2026년 1월 5일

[논문리뷰] KV-Embedding: Training-free Text Embedding via Internal KV Re-routing in Decoder-only LLMs

Yi Yang이 arXiv에 게시한 'KV-Embedding: Training-free Text Embedding via Internal KV Re-routing in Decoder-only LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Text Embedding #Decoder-only LLMs #Training-free #KV Re-routing #Causal Attention #Representation Learning #Intrinsic Dimensionality

2026년 1월 5일

[논문리뷰] K-EXAONE Technical Report

arXiv에 게시된 'K-EXAONE Technical Report' 논문에 대한 자세한 리뷰입니다.

#Review #Multilingual Language Model #Mixture-of-Experts (MoE)#Long Context #AI Safety #Korean AI #Foundation Model #Reinforcement Learning (RL)

2026년 1월 5일

[논문리뷰] InfiniteVGGT: Visual Geometry Grounded Transformer for Endless Streams

arXiv에 게시된 'InfiniteVGGT: Visual Geometry Grounded Transformer for Endless Streams' 논문에 대한 자세한 리뷰입니다.

#Review #3D Reconstruction #Transformer #Streaming Perception #Memory Management #KV Cache Pruning #Visual Geometry #Temporal Consistency #Continuous Learning

2026년 1월 5일

[논문리뷰] IMA++: ISIC Archive Multi-Annotator Dermoscopic Skin Lesion Segmentation Dataset

arXiv에 게시된 'IMA++: ISIC Archive Multi-Annotator Dermoscopic Skin Lesion Segmentation Dataset' 논문에 대한 자세한 리뷰입니다.

#Review #Dermoscopy #Skin Lesion Segmentation #Multi-Annotator Dataset #Inter-Annotator Variability #ISIC Archive #Medical Image Analysis #Machine Learning #Data Annotation

2026년 1월 5일

[논문리뷰] GARDO: Reinforcing Diffusion Models without Reward Hacking

Zhiyong Wang이 arXiv에 게시한 'GARDO: Reinforcing Diffusion Models without Reward Hacking' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion Models #Reinforcement Learning #Reward Hacking #KL Regularization #Adaptive Regularization #Diversity Optimization #Text-to-Image Generation

2026년 1월 5일

[논문리뷰] Falcon-H1R: Pushing the Reasoning Frontiers with a Hybrid Model for Efficient Test-Time Scaling

arXiv에 게시된 'Falcon-H1R: Pushing the Reasoning Frontiers with a Hybrid Model for Efficient Test-Time Scaling' 논문에 대한 자세한 리뷰입니다.

#Review #Reasoning #Small Language Models (SLMs)#Hybrid Architecture #Test-Time Scaling (TTS)#Supervised Fine-Tuning (SFT)#Reinforcement Learning (RL)#DeepConf #Computational Efficiency

2026년 1월 5일

[논문리뷰] DreamID-V:Bridging the Image-to-Video Gap for High-Fidelity Face Swapping via Diffusion Transformer

arXiv에 게시된 'DreamID-V:Bridging the Image-to-Video Gap for High-Fidelity Face Swapping via Diffusion Transformer' 논문에 대한 자세한 리뷰입니다.

#Review #Video Face Swapping #Diffusion Transformer #Identity Preservation #Temporal Consistency #Modality-Aware Conditioning #Reinforcement Learning #Data Synthesis

2026년 1월 5일

[논문리뷰] Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits

arXiv에 게시된 'Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Self-Awareness #Failure Prediction #Internal States #Attention Mechanisms #Neural Network Probes #Computational Efficiency #Zero-Shot Transfer

2026년 1월 5일

[논문리뷰] COMPASS: A Framework for Evaluating Organization-Specific Policy Alignment in LLMs

arXiv에 게시된 'COMPASS: A Framework for Evaluating Organization-Specific Policy Alignment in LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Evaluation #Policy Alignment #Organizational Policies #AI Safety #Adversarial Robustness #Refusal Behavior #Prompt Engineering #Fine-tuning

2026년 1월 5일

[논문리뷰] Youtu-Agent: Scaling Agent Productivity with Automated Generation and Hybrid Policy Optimization

arXiv에 게시된 'Youtu-Agent: Scaling Agent Productivity with Automated Generation and Hybrid Policy Optimization' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Agents #Automated Agent Generation #Reinforcement Learning #Hybrid Policy Optimization #Tool Synthesis #In-context Learning #Agent Framework #Scalability

2026년 1월 4일

[논문리뷰] Taming Hallucinations: Boosting MLLMs' Video Understanding via Counterfactual Video Generation

arXiv에 게시된 'Taming Hallucinations: Boosting MLLMs' Video Understanding via Counterfactual Video Generation' 논문에 대한 자세한 리뷰입니다.

#Review #MLLMs #Video Understanding #Hallucinations #Counterfactual Generation #Diffusion Models #Reinforcement Learning #QA Dataset #DNA-Train

2026년 1월 4일

[논문리뷰] SenseNova-MARS: Empowering Multimodal Agentic Reasoning and Search via Reinforcement Learning

arXiv에 게시된 'SenseNova-MARS: Empowering Multimodal Agentic Reasoning and Search via Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Agents #Reinforcement Learning #Vision-Language Models #Tool Use #Agentic Reasoning #Image Search #HR-MMSearch #BN-GSPO

2026년 1월 4일

[논문리뷰] Nested Learning: The Illusion of Deep Learning Architectures

Vahab Mirrokni이 arXiv에 게시한 'Nested Learning: The Illusion of Deep Learning Architectures' 논문에 대한 자세한 리뷰입니다.

#Review #Nested Learning #Continual Learning #In-context Learning #Associative Memory #Multi-Timescale Memory #Self-Modifying Models #Optimizers

2026년 1월 4일

[논문리뷰] NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos

Feng Wang이 arXiv에 게시한 'NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos' 논문에 대한 자세한 리뷰입니다.

#Review #4D World Model #Gaussian Splatting #Monocular Video #Novel View Synthesis #Video Generation #Feed-Forward Reconstruction #Degradation Simulation

2026년 1월 4일

[논문리뷰] MorphAny3D: Unleashing the Power of Structured Latent in 3D Morphing

Jian Yang이 arXiv에 게시한 'MorphAny3D: Unleashing the Power of Structured Latent in 3D Morphing' 논문에 대한 자세한 리뷰입니다.

#Review #3D Morphing #Structured Latent (SLAT)#Generative Models #Attention Mechanisms #Training-Free Framework #Cross-Category Transitions #Temporal Coherence

2026년 1월 4일

[논문리뷰] InfoSynth: Information-Guided Benchmark Synthesis for LLMs

arXiv에 게시된 'InfoSynth: Information-Guided Benchmark Synthesis for LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Benchmark Synthesis #LLM Evaluation #Code Generation #Information Theory #Genetic Algorithms #Novelty Metrics #Diversity Metrics

2026년 1월 4일