北京通用人工智能研究院BIGAI

论文索引

AdaptMMBench: Benchmarking Adaptive Multimodal Reasoning for Mode Selection and Reasoning Process

Fully Decoupled Reinforcement Learning with Adaptive Data Curation for GUI Agents

EgoProx: Evaluating MLLMs on Egocentric 3D Proximity Reasoning Across a Cognitive Hierarchy

Human Intelligence-Inspired Testing for the Developmental Stages of Artificial General Intelligence: From General to Applicable

KORE: Enhancing Knowledge Injection for Large Multimodal Models via Knowledge-Oriented Augmentations and Constraints

DecomVoxel: Harnessing 3D-Native Priors with Guided In-situ Denoising Optimization for Decompositional Scene Reconstruction