Artigos de Pesquisa em IA Diários

Mask^2DiT: Dual Mask-based Diffusion Transformer for Multi-Scene Long Video Generation

DINeMo: Learning Neural Mesh Models with no 3D Annotations

Qwen2.5-Omni Technical Report

LogQuant: Log-Distributed 2-Bit Quantization of KV Cache with Superior Accuracy Preservation

PathoHR: Breast Cancer Survival Prediction on High-Resolution Pathological Images

Open Deep Search: Democratizing Search with Open-source Reasoning Agents

ViLBench: A Suite for Vision-Language Process Reward Modeling

UniHDSA: A Unified Relation Prediction Approach for Hierarchical Document Structure Analysis

Trajectory Balance with Asynchrony: Decoupling Exploration and Learning for Fast, Scalable LLM Post-Training

GenHancer: Imperfect Generative Models are Secretly Strong Vision-Centric Enhancers

Image as an IMU: Estimating Camera Motion from a Single Motion-Blurred Image

LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?

RecTable: Fast Modeling Tabular Data with Rectified Flow

Unconditional Priors Matter! Improving Conditional Generation of Fine-Tuned Diffusion Models

Beyond Words: Advancing Long-Text Image Generation via Multimodal Autoregressive Models

AccVideo: Accelerating Video Diffusion Model with Synthetic Dataset

Attention IoU: Examining Biases in CelebA using Attention Maps

MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search

Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals

Wan: Open and Advanced Large-Scale Video Generative Models

Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs

ADS-Edit: A Multimodal Knowledge Editing Dataset for Autonomous Driving Systems

Gemma 3 Technical Report

BizGen: Advancing Article-level Visual Text Rendering for Infographics Generation

Gemini Robotics: Bringing AI into the Physical World

Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy

RONA: Pragmatically Diverse Image Captioning with Coherence Relations

Unlocking Efficient Long-to-Short LLM Reasoning with Model Merging