Articoli di Ricerca IA Giornalieri

RWKV-7 "Goose" with Expressive Dynamic State Evolution

DAPO: An Open-Source LLM Reinforcement Learning System at Scale

Impossible Videos

Creation-MMBench: Assessing Context-Aware Creative Intelligence in MLLM

Infinite Mobility: Scalable High-Fidelity Synthesis of Articulated Objects via Procedural Generation

DeepPerception: Advancing R1-like Cognitive Visual Perception in MLLMs for Knowledge-Intensive Visual Grounding

AudioX: Diffusion Transformer for Anything-to-Audio Generation

Aligning Multimodal LLM with Human Preference: A Survey

CapArena: Benchmarking and Analyzing Detailed Image Captioning in the LLM Era

Frac-Connections: Fractional Extension of Hyper-Connections

Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control

FlexWorld: Progressively Expanding 3D Scenes for Flexiable-View Synthesis

Atlas: Multi-Scale Attention Improves Long Context Image Modeling

Measuring AI Ability to Complete Long Tasks

Temporal Consistency for LLM Reasoning Process Error Identification

Concat-ID: Towards Universal Identity-Preserving Video Synthesis

MPBench: A Comprehensive Multimodal Reasoning Benchmark for Process Errors Identification

Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection

MM-Spatial: Exploring 3D Spatial Understanding in Multimodal LLMs

Towards Self-Improving Systematic Cognition for Next-Generation Foundation MLLMs

Florenz: Scaling Laws for Systematic Generalization in Vision-Language Models

Pensez: Less Data, Better Reasoning -- Rethinking French LLM

PEBench: A Fictitious Dataset to Benchmark Machine Unlearning for Multimodal Large Language Models

EvalTree: Profiling Language Model Weaknesses via Hierarchical Capability Trees

Learning to Inference Adaptively for Multimodal Large Language Models

PyGDA: A Python Library for Graph Domain Adaptation

MeshFleet: Filtered and Annotated 3D Vehicle Dataset for Domain Specific Generative Modeling

Hyperbolic Safety-Aware Vision-Language Models

KUDA: Keypoints to Unify Dynamics Learning and Visual Prompting for Open-Vocabulary Robotic Manipulation

RoCo-Sim: Enhancing Roadside Collaborative Perception through Foreground Simulation

CoLMDriver: LLM-based Negotiation Benefits Cooperative Autonomous Driving