AI研究論文每日精選

BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset

DeCLIP: Decoupled Learning for Open-Vocabulary Dense Perception

Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures

Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysis

UniSkill: Imitating Human Videos via Cross-Embodiment Skill Representations

SweRank: Software Issue Localization with Code Ranking

CAST: Component-Aligned 3D Scene Reconstruction from an RGB Image

WavReward: Spoken Dialogue Models With Generalist Reward Evaluators

Omni-R1: Do You Really Need Audio to Fine-Tune Your Audio LLM?

VCRBench: Exploring Long-form Causal Reasoning Capabilities of Large Video Language Models

DetReIDX: A Stress-Test Dataset for Real-World UAV-Based Person Recognition

Visually Interpretable Subtask Reasoning for Visual Question Answering

LightLab: Controlling Light Sources in Images with Diffusion Models

Behind Maya: Building a Multilingual Vision Language Model

Understanding and Mitigating Toxicity in Image-Text Pretraining Datasets: A Case Study on LLaVA

Steepest Descent Density Control for Compact 3D Gaussian Splatting