Review

[논문리뷰] Gradient-Attention Guided Dual-Masking Synergetic Framework for Robust Text-based Person Retrieval

Kaicheng Yang이 arXiv에 게시한 'Gradient-Attention Guided Dual-Masking Synergetic Framework for Robust Text-based Person Retrieval' 논문에 대한 자세한 리뷰입니다.

#Review #Text-based Person Retrieval #CLIP #MLLM #Data Curation #Dual-Masking #Gradient-Attention #WebPerson Dataset

2025년 9월 12일

[논문리뷰] FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmark

Shuai Bai이 arXiv에 게시한 'FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmark' 논문에 대한 자세한 리뷰입니다.

#Review #Text-to-Image Generation #Reasoning Dataset #Benchmark #Generation Chain-of-Thought #Vision-Language Model #Image Aesthetics #Prompt Alignment

2025년 9월 12일

[논문리뷰] EchoX: Towards Mitigating Acoustic-Semantic Gap via Echo Training for Speech-to-Speech LLMs

Kaiqi Kou이 arXiv에 게시한 'EchoX: Towards Mitigating Acoustic-Semantic Gap via Echo Training for Speech-to-Speech LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Speech-to-Speech LLMs #Acoustic-Semantic Gap #Echo Training #Unit Language #Streaming Inference #Knowledge-based QA

2025년 9월 12일

[논문리뷰] Can Understanding and Generation Truly Benefit Together -- or Just Coexist?

Hui Han이 arXiv에 게시한 'Can Understanding and Generation Truly Benefit Together -- or Just Coexist?' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal Understanding #Multimodal Generation #Unified Models #Auto-Encoder #Reinforcement Learning #Image-to-Text #Text-to-Image #Reconstruction Fidelity

2025년 9월 12일

[논문리뷰] 2D Gaussian Splatting with Semantic Alignment for Image Inpainting

Guangming Lu이 arXiv에 게시한 '2D Gaussian Splatting with Semantic Alignment for Image Inpainting' 논문에 대한 자세한 리뷰입니다.

#Review #Image Inpainting #2D Gaussian Splatting #Semantic Alignment #DINO Features #Patch-level Rasterization #Continuous Representation #Generative Models

2025년 9월 12일

[논문리뷰] <think> So let's replace this phrase with insult... </think> Lessons learned from generation of toxic texts with LLMs

Alexander Panchenko이 arXiv에 게시한 '<think> So let's replace this phrase with insult... </think> Lessons learned from generation of toxic texts with LLMs' 논문에 대한 자세한 리뷰입니다.

#Review #Toxic Text Generation #LLMs #Text Detoxification #Lexical Diversity #Synthetic Data #Human Annotation #Style Transfer

2025년 9월 11일

[논문리뷰] RewardDance: Reward Scaling in Visual Generation

Liang Li이 arXiv에 게시한 'RewardDance: Reward Scaling in Visual Generation' 논문에 대한 자세한 리뷰입니다.

#Review #Reward Model #Visual Generation #RLHF #VLM #Reward Scaling #Reward Hacking #Generative Paradigm #Context Scaling #Text-to-Image #Text-to-Video

2025년 9월 11일

[논문리뷰] P3-SAM: Native 3D Part Segmentation

Yunhan Yang이 arXiv에 게시한 'P3-SAM: Native 3D Part Segmentation' 논문에 대한 자세한 리뷰입니다.

#Review #3D Part Segmentation #Point Cloud Segmentation #Prompt-based Segmentation #Deep Learning #Transformer #Interactive Segmentation #Automatic Segmentation #Native 3D

2025년 9월 11일

[논문리뷰] Hunyuan-MT Technical Report

Yang Du이 arXiv에 게시한 'Hunyuan-MT Technical Report' 논문에 대한 자세한 리뷰입니다.

#Review #Machine Translation #Large Language Model #Multilingual #Low-Resource Languages #Reinforcement Learning #Weak-to-Strong Learning #Slow Thinking

2025년 9월 11일

[논문리뷰] HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants

Jacy Reese Anthis이 arXiv에 게시한 'HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants' 논문에 대한 자세한 리뷰입니다.

#Review #Human Agency #AI Assistants #LLM Evaluation #Benchmark #Sociotechnical AI #AI Alignment #Scalable Evaluation

2025년 9월 11일

[논문리뷰] EnvX: Agentize Everything with Agentic AI

Wenzheng Tom Tang이 arXiv에 게시한 'EnvX: Agentize Everything with Agentic AI' 논문에 대한 자세한 리뷰입니다.

#Review #Agentic AI #Multi-Agent Systems #Code Repository #Agentization #Natural Language Interaction #Agent-to-Agent Protocol #LLM-based Agents

2025년 9월 11일

[논문리뷰] AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning

Honglin Guo이 arXiv에 게시한 'AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Agents #Reinforcement Learning #Multi-Turn Interaction #Long-Horizon Decision Making #Agent Framework #Exploration-Exploitation #Progressive Scaling

2025년 9월 11일

[논문리뷰] A Survey of Reinforcement Learning for Large Reasoning Models

Runze Liu이 arXiv에 게시한 'A Survey of Reinforcement Learning for Large Reasoning Models' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #Large Reasoning Models #LLMs #Reward Design #Policy Optimization #Verifiable Rewards #Agentic AI #Multimodal AI

2025년 9월 11일

[논문리뷰] 3D and 4D World Modeling: A Survey

Ao Liang이 arXiv에 게시한 '3D and 4D World Modeling: A Survey' 논문에 대한 자세한 리뷰입니다.

#Review #3D World Modeling #4D World Modeling #Generative Models #Predictive Models #LiDAR #Occupancy Grids #Video Generation #Autonomous Driving #Robotics

2025년 9월 11일

[논문리뷰] ΔL Normalization: Rethink Loss Aggregation in RLVR

Lili Qiu이 arXiv에 게시한 'ΔL Normalization: Rethink Loss Aggregation in RLVR' 논문에 대한 자세한 리뷰입니다.

#Review #Reinforcement Learning #LLMs #Gradient Variance #Loss Aggregation #Unbiased Estimator #RLVR #Policy Gradient #Normalization

2025년 9월 10일

[논문리뷰] Visual Representation Alignment for Multimodal Large Language Models

Heeseong Shin이 arXiv에 게시한 'Visual Representation Alignment for Multimodal Large Language Models' 논문에 대한 자세한 리뷰입니다.

#Review #Multimodal LLMs #Visual Representation Alignment #Foundation Models #Regularization #Fine-grained Visual Understanding #Spatial Reasoning #Object Counting #Vision-Language Models

2025년 9월 10일

[논문리뷰] UMO: Scaling Multi-Identity Consistency for Image Customization via Matching Reward

Fei Ding이 arXiv에 게시한 'UMO: Scaling Multi-Identity Consistency for Image Customization via Matching Reward' 논문에 대한 자세한 리뷰입니다.

#Review #Image Customization #Multi-Identity Generation #Identity Consistency #Identity Confusion #Reinforcement Learning #Diffusion Models #Matching Reward #Global Assignment

2025년 9월 10일

[논문리뷰] Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding

Yongcheng Zeng이 arXiv에 게시한 'Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding' 논문에 대한 자세한 리뷰입니다.

#Review #RLVR #LLM Reasoning #Adaptive Learning #Hint Scaffolding #Item Response Theory #Exploration Efficiency #Problem Difficulty #Policy Optimization

2025년 9월 10일

[논문리뷰] SimpleQA Verified: A Reliable Factuality Benchmark to Measure Parametric Knowledge

Dipanjan Das이 arXiv에 게시한 'SimpleQA Verified: A Reliable Factuality Benchmark to Measure Parametric Knowledge' 논문에 대한 자세한 리뷰입니다.

#Review #LLM Factuality #Parametric Knowledge #Benchmark #Question Answering #Data Curation #Evaluation Metrics #Hallucination Mitigation #Large Language Models

2025년 9월 10일

[논문리뷰] Reconstruction Alignment Improves Unified Multimodal Models

XuDong Wang이 arXiv에 게시한 'Reconstruction Alignment Improves Unified Multimodal Models' 논문에 대한 자세한 리뷰입니다.

#Review #Unified Multimodal Models #Image Generation #Image Editing #Post-training #Self-supervised Learning #Reconstruction Alignment #Visual Embeddings

2025년 9월 10일