Artigos de Pesquisa em IA Diários

Generalizable Humanoid Manipulation with Improved 3D Diffusion Policies

Semantic Image Inversion and Editing using Rectified Stochastic Differential Equations

Toward General Instruction-Following Alignment for Retrieval-Augmented Generation

Latent Action Pretraining from Videos

LOKI: A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Models

MEGA-Bench: Scaling Multimodal Evaluation to over 500 Real-World Tasks

MMCOMPOSITION: Revisiting the Compositionality of Pre-trained Vision-Language Models

VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents

Tree of Problems: Improving structured problem solving with compositionality

Rethinking Data Selection at Scale: Random Selection is Almost All You Need

ReLU's Revival: On the Entropic Overload in Normalization-Free Large Language Models

LiveXiv -- A Multi-Modal Live Benchmark Based on Arxiv Papers Content

Thinking LLMs: General Instruction Following with Thought Generation

LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory

Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models

Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attention

DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads

Animate-X: Universal Character Image Animation with Enhanced Motion Representation

MMIE: Massive Multimodal Interleaved Comprehension Benchmark for Large Vision-Language Models

The Same But Different: Structural Similarities and Differences in Multilingual Language Modeling

TVBench: Redesigning Video-Language Evaluation

TemporalBench: Benchmarking Fine-grained Temporal Understanding for Multimodal Video Models