Review

[논문리뷰] Do What I Say: A Spoken Prompt Dataset for Instruction-Following

Marek Kasztelnik이 arXiv에 게시한 'Do What I Say: A Spoken Prompt Dataset for Instruction-Following' 논문에 대한 자세한 리뷰입니다.

#Review #Speech Language Models #Instruction Following #Multilingual Dataset #Spoken Prompts #Benchmark #SLLM Evaluation #Prompt Diversity

2026년 3월 10일

[논문리뷰] Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards

arXiv에 게시된 'Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #LLM Calibration #Over-confidence #Decoupled Optimization #Verifiable Rewards #Policy Optimization #Expected Calibration Error

2026년 3월 10일

[논문리뷰] ConFu: Contemplate the Future for Better Speculative Sampling

arXiv에 게시된 'ConFu: Contemplate the Future for Better Speculative Sampling' 논문에 대한 자세한 리뷰입니다.

#Review #Speculative Decoding #LLM Inference Acceleration #Draft Model #Future Prediction #Contemplate Tokens #Mixture-of-Experts #Token Acceptance Rate #Speedup Ratio

2026년 3월 10일

[논문리뷰] Compiler-First State Space Duality and Portable O(1) Autoregressive Caching for Inference

arXiv에 게시된 'Compiler-First State Space Duality and Portable O(1) Autoregressive Caching for Inference' 논문에 대한 자세한 리뷰입니다.

#Review #State Space Models #Mamba-2 #XLA #JAX #Compiler Codegen #Autoregressive Caching #Hardware Portability #Inference Optimization

2026년 3월 10일

[논문리뷰] BrandFusion: A Multi-Agent Framework for Seamless Brand Integration in Text-to-Video Generation

arXiv에 게시된 'BrandFusion: A Multi-Agent Framework for Seamless Brand Integration in Text-to-Video Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Text-to-Video Generation #Multi-Agent System #Brand Integration #Prompt Engineering #Large Language Models (LLMs)#LoRA Fine-tuning #Contextual Adaptation

2026년 3월 10일

[논문리뷰] Are Audio-Language Models Listening? Audio-Specialist Heads for Adaptive Audio Steering

arXiv에 게시된 'Are Audio-Language Models Listening? Audio-Specialist Heads for Adaptive Audio Steering' 논문에 대한 자세한 리뷰입니다.

#Review #Audio-Language Models (LALMs)#Text Dominance #Mechanistic Interpretability #Attention Heads #Activation Steering #Multimodal Grounding #Inference-time Intervention

2026년 3월 10일

[논문리뷰] Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training

arXiv에 게시된 'Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training' 논문에 대한 자세한 리뷰입니다.

#Review #Financial LLMs #Data-Centric AI #Distillation #Chain-of-Thought (CoT)#Reinforcement Learning (RL)#Supervised Fine-Tuning (SFT)#Difficulty-Aware Training #Data Quality

2026년 3월 9일

[논문리뷰] TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Reward

arXiv에 게시된 'TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Reward' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion Models #Reinforcement Learning #Non-Differentiable Rewards #Few-Step Generation #Trajectory Distribution Matching #Surrogate Reward Learning #Text-to-Image

2026년 3월 9일

[논문리뷰] Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs

arXiv에 게시된 'Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion LLMs #Autoregressive LLMs #Representational Analysis #Layer Skipping #Inference Efficiency #Initialization Bias #FLOPs Reduction #Recency Bias

2026년 3월 9일

[논문리뷰] Scale Space Diffusion

Abhinav Shrivastava이 arXiv에 게시한 'Scale Space Diffusion' 논문에 대한 자세한 리뷰입니다.

#Review #Diffusion Models #Scale Space Theory #Generative Models #Multi-resolution Image Generation #UNet Architecture #Image Upsampling #Non-Isotropic Noise

2026년 3월 9일

[논문리뷰] PureCC: Pure Learning for Text-to-Image Concept Customization

arXiv에 게시된 'PureCC: Pure Learning for Text-to-Image Concept Customization' 논문에 대한 자세한 리뷰입니다.

#Review #Text-to-Image #Concept Customization #Flow-based Models #Pure Learning #Model Preservation #Adaptive Guidance #LoRA

2026년 3월 9일

[논문리뷰] PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents

Hongsheng Li이 arXiv에 게시한 'PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents' 논문에 대한 자세한 리뷰입니다.

#Review #Proactive Agents #GUI Automation #Intent Recommendation #Multimodal LLMs #Benchmark #Memory-aware Framework #Human-Computer Interaction

2026년 3월 9일

[논문리뷰] $OneMillion-Bench: How Far are Language Agents from Human Experts?

arXiv에 게시된 '$OneMillion-Bench: How Far are Language Agents from Human Experts?' 논문에 대한 자세한 리뷰입니다.

#Review #Language Agents #Benchmarking #Expert Evaluation #Economic Value #Professional Tasks #Rubric-based Evaluation #Multi-step Reasoning #Reliability #Domain Adaptation

2026년 3월 9일

[논문리뷰] NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Driving

arXiv에 게시된 'NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Driving' 논문에 대한 자세한 리뷰입니다.

#Review #Autonomous Driving #Vision-Language Models #Motion Planning #High-Level Reasoning #Decoupled Architecture #Supervised Fine-tuning #NuScenes Benchmark

2026년 3월 9일

[논문리뷰] NLE: Non-autoregressive LLM-based ASR by Transcript Editing

arXiv에 게시된 'NLE: Non-autoregressive LLM-based ASR by Transcript Editing' 논문에 대한 자세한 리뷰입니다.

#Review #Non-Autoregressive ASR #LLM-based ASR #Transcript Editing #CTC #Transformer #LoRA #Real-time ASR #Inference Speed

2026년 3월 9일

[논문리뷰] Lost in Stories: Consistency Bugs in Long Story Generation by LLMs

Hongzhi Li이 arXiv에 게시한 'Lost in Stories: Consistency Bugs in Long Story Generation by LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Large Language Models (LLMs)#Story Generation #Narrative Consistency #Benchmark #Automated Evaluation #Error Analysis #Long-Form Text Generation #Consistency Error Density (CED)

2026년 3월 9일

[논문리뷰] LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory

arXiv에 게시된 'LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory' 논문에 대한 자세한 리뷰입니다.

#Review #3D Reconstruction #Long-Context #Hybrid Memory #Sliding Window Attention (SWA)#Test-Time Training (TTT)#Transformer #Visual SLAM #Sequence Modeling

2026년 3월 9일

[논문리뷰] How Far Can Unsupervised RLVR Scale LLM Training?

Shangziqi Zhao이 arXiv에 게시한 'How Far Can Unsupervised RLVR Scale LLM Training?' 논문에 대한 자세한 리뷰입니다.

#Review #Unsupervised Reinforcement Learning #LLM Training #Intrinsic Rewards #External Rewards #Model Collapse #RLVR #Model Prior #Self-Verification

2026년 3월 9일

[논문리뷰] Holi-Spatial: Evolving Video Streams into Holistic 3D Spatial Intelligence

Yuning Gong이 arXiv에 게시한 'Holi-Spatial: Evolving Video Streams into Holistic 3D Spatial Intelligence' 논문에 대한 자세한 리뷰입니다.

#Review #3D Spatial Intelligence #Video Stream Processing #Automated Data Curation #3D Gaussian Splatting (3DGS)#Vision-Language Models (VLMs)#Open-Vocabulary Segmentation #Spatial Reasoning #Multimodal Datasets

2026년 3월 9일

[논문리뷰] HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoising

Bin Liu이 arXiv에 게시한 'HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoising' 논문에 대한 자세한 리뷰입니다.

#Review #Autoregressive Video Generation #Hierarchical Denoising #Diffusion Models #Temporal Continuity #Error Propagation #Forward-KL Regularization #Long Video Synthesis #Pipelined Parallelism

2026년 3월 9일