Review

[논문리뷰] Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation

Han Shi이 arXiv에 게시한 'Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Autoregressive Models #Text-to-Image Generation #Inference Acceleration #Jacobi Decoding #Denoising Diffusion Models #Speculative Decoding #Multi-token Prediction #Fine-tuning

2025년 10월 13일

[논문리뷰] SpaceVista: All-Scale Visual Spatial Reasoning from mm to km

Kaituo Feng이 arXiv에 게시한 'SpaceVista: All-Scale Visual Spatial Reasoning from mm to km' 논문에 대한 자세한 리뷰입니다.

#Review #Spatial Reasoning #Multi-Scale Vision #MLLM #Dataset #Scale Experts #Reinforcement Learning #Computer Vision #Robotics

2025년 10월 13일

[논문리뷰] ReviewerToo: Should AI Join The Program Committee? A Look At The Future of Peer Review

Christopher Pal이 arXiv에 게시한 'ReviewerToo: Should AI Join The Program Committee? A Look At The Future of Peer Review' 논문에 대한 자세한 리뷰입니다.

#Review #Peer Review #AI-Assisted Review #Large Language Models #LLM Agents #Meta-Review #Conference Submissions #Reviewer Personas #Evaluation Metrics

2025년 10월 13일

[논문리뷰] R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?

arXiv에 게시된 'R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?' 논문에 대한 자세한 리뷰입니다.

#Review #Long-Horizon Reasoning #Query Composition #Large Reasoning Models #Reinforcement Learning #Benchmark Evaluation #Thinking Budget #Performance Degradation #Chain-of-Thought

2025년 10월 13일

[논문리뷰] Pseudo2Real: Task Arithmetic for Pseudo-Label Correction in Automatic Speech Recognition

Shang-Tse Chen이 arXiv에 게시한 'Pseudo2Real: Task Arithmetic for Pseudo-Label Correction in Automatic Speech Recognition' 논문에 대한 자세한 리뷰입니다.

#Review #ASR #Pseudo-labeling #Domain Adaptation #Task Arithmetic #Correction Vector #Accent Adaptation #Speaker Clustering #Model Editing

2025년 10월 13일

[논문리뷰] Progressive Gaussian Transformer with Anisotropy-aware Sampling for Open Vocabulary Occupancy Prediction

danxuhk이 arXiv에 게시한 'Progressive Gaussian Transformer with Anisotropy-aware Sampling for Open Vocabulary Occupancy Prediction' 논문에 대한 자세한 리뷰입니다.

#Review #3D Occupancy Prediction #Open Vocabulary #Gaussian Splatting #Transformer #Progressive Densification #Anisotropy-aware Sampling #Autonomous Driving

2025년 10월 13일

[논문리뷰] PhysToolBench: Benchmarking Physical Tool Understanding for MLLMs

Xu Zheng이 arXiv에 게시한 'PhysToolBench: Benchmarking Physical Tool Understanding for MLLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Large Language Models (MLLMs)#Physical Tool Understanding #Benchmarking #Embodied AI #Visual Question Answering (VQA)#Tool Affordances #Reasoning

2025년 10월 13일

[논문리뷰] Parallel Test-Time Scaling for Latent Reasoning Models

arXiv에 게시된 'Parallel Test-Time Scaling for Latent Reasoning Models' 논문에 대한 자세한 리뷰입니다.

#Review #Latent Reasoning #Test-Time Scaling #Parallel Inference #Stochastic Sampling #Monte Carlo Dropout #Additive Gaussian Noise #Latent Reward Model #Trajectory Aggregation

2025년 10월 13일

[논문리뷰] One Patch to Caption Them All: A Unified Zero-Shot Captioning Framework

Giuseppe Amato이 arXiv에 게시한 'One Patch to Caption Them All: A Unified Zero-Shot Captioning Framework' 논문에 대한 자세한 리뷰입니다.

#Review #Zero-Shot Captioning #Region-Level Captioning #Vision Transformers #DINOv2 #Patch-Centric #Modality Gap Mitigation #Visual-Language Models

2025년 10월 13일

[논문리뷰] Multimodal Prompt Optimization: Why Not Leverage Multiple Modalities for MLLMs

arXiv에 게시된 'Multimodal Prompt Optimization: Why Not Leverage Multiple Modalities for MLLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal AI #Prompt Optimization #MLLMs #Bayesian Optimization #Cross-modal Alignment #Prompt Engineering #Generative AI #Exploration-Exploitation

2025년 10월 13일

[논문리뷰] Mitigating Overthinking through Reasoning Shaping

Wen Luo이 arXiv에 게시한 'Mitigating Overthinking through Reasoning Shaping' 논문에 대한 자세한 리뷰입니다.

#Review #Large Reasoning Models (LRMs)#RLVR #Overthinking Mitigation #Reasoning Shaping #Segment-level Penalization #Computational Efficiency #Training Stability #Length-aware Weighting

2025년 10월 13일

[논문리뷰] MRMR: A Realistic and Expert-Level Multidisciplinary Benchmark for Reasoning-Intensive Multimodal Retrieval

Tingyu Song이 arXiv에 게시한 'MRMR: A Realistic and Expert-Level Multidisciplinary Benchmark for Reasoning-Intensive Multimodal Retrieval' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Retrieval #Benchmark #Reasoning #Multidisciplinary #Expert-Level #Image-Text Interleaving #Contradiction Retrieval

2025년 10월 13일

[논문리뷰] KORMo: Korean Open Reasoning Model for Everyone

arXiv에 게시된 'KORMo: Korean Open Reasoning Model for Everyone' 논문에 대한 자세한 리뷰입니다.

#Review #Large Language Model #Korean #Bilingual #Synthetic Data #Fully Open Model #Tokenizer #Reasoning #Pretraining #Instruction Tuning

2025년 10월 13일

[논문리뷰] Instant4D: 4D Gaussian Splatting in Minutes

Li Lu이 arXiv에 게시한 'Instant4D: 4D Gaussian Splatting in Minutes' 논문에 대한 자세한 리뷰입니다.

#Review #4D Gaussian Splatting #Dynamic View Synthesis #Monocular Reconstruction #Visual SLAM #Grid Pruning #Real-time Rendering #GPU Memory Optimization

2025년 10월 13일

[논문리뷰] Hybrid-grained Feature Aggregation with Coarse-to-fine Language Guidance for Self-supervised Monocular Depth Estimation

Zekun Qi이 arXiv에 게시한 'Hybrid-grained Feature Aggregation with Coarse-to-fine Language Guidance for Self-supervised Monocular Depth Estimation' 논문에 대한 자세한 리뷰입니다.

#Review #Self-supervised Monocular Depth Estimation #Foundation Models #CLIP #DINO #Language Guidance #Coarse-to-fine Learning #Feature Aggregation #3D Perception

2025년 10월 13일

[논문리뷰] GTAlign: Game-Theoretic Alignment of LLM Assistants for Mutual Welfare

arXiv에 게시된 'GTAlign: Game-Theoretic Alignment of LLM Assistants for Mutual Welfare' 논문에 대한 자세한 리뷰입니다.

#Review #Large Language Models #LLM Alignment #Game Theory #Reinforcement Learning #Mutual Welfare #Payoff Matrix #Strategic Decision Making #Human-AI Interaction

2025년 10월 13일

[논문리뷰] Dyna-Mind: Learning to Simulate from Experience for Better AI Agents

Qianhui Wu이 arXiv에 게시한 'Dyna-Mind: Learning to Simulate from Experience for Better AI Agents' 논문에 대한 자세한 리뷰입니다.

#Review #AI Agents #Reinforcement Learning #World Models #Simulation #Reasoning #Language Models #Planning #Interactive AI

2025년 10월 13일

[논문리뷰] Don't Waste Mistakes: Leveraging Negative RL-Groups via Confidence Reweighting

Julia Kempe이 arXiv에 게시한 'Don't Waste Mistakes: Leveraging Negative RL-Groups via Confidence Reweighting' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #Large Language Models #Reasoning Tasks #GRPO #Negative Samples #Reward Modeling #Confidence Reweighting #Mathematical Reasoning

2025년 10월 13일

[논문리뷰] DISCO: Diversifying Sample Condensation for Efficient Model Evaluation

arXiv에 게시된 'DISCO: Diversifying Sample Condensation for Efficient Model Evaluation' 논문에 대한 자세한 리뷰입니다.

#Review #Efficient Evaluation #Sample Condensation #Model Disagreement #Predictive Diversity #Performance Prediction #Large Language Models #Model Signatures #Meta-modeling

2025년 10월 13일

[논문리뷰] D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI

Haebin Seong이 arXiv에 게시한 'D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI' 논문에 대한 자세한 리뷰입니다.

#Review #Embodied AI #Vision-Action Pretraining #Desktop Data #Inverse Dynamics Model (IDM)#Pseudo-labeling #Robotics #Generalization #Data Compression

2025년 10월 13일