AI Onderzoekspapers Dagelijks

Agent-R: Training Language Model Agents to Reflect via Iterative Self-Training

MMVU: Measuring Expert-Level Multi-Discipline Video Understanding

Demons in the Detail: On Implementing Load Balancing Loss for Training Specialized Mixture-of-Expert Models

UI-TARS: Pioneering Automated GUI Interaction with Native Agents

TokenVerse: Versatile Multi-concept Personalization in Token Modulation Space

InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model

Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation

Reasoning Language Models: A Blueprint

Mobile-Agent-E: Self-Evolving Mobile Assistant for Complex Tasks

Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

Learn-by-interact: A Data-Centric Framework for Self-Adaptive Agents in Realistic Environments

Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise

Condor: Enhance LLM Alignment with Knowledge-Driven Data Synthesis and Refinement

EMO2: End-Effector Guided Audio-Driven Avatar Video Generation

GPS as a Control Signal for Image Generation

Taming Teacher Forcing for Masked Autoregressive Video Generation

The Geometry of Tokens in Internal Representations of Large Language Models

MSTS: A Multimodal Safety Test Suite for Vision-Language Models

Fixing Imbalanced Attention to Mitigate In-Context Hallucination of Large Vision-Language Model

Panoramic Interests: Stylistic-Content Aware Personalized Headline Generation