Fine-tuning visual models has been widely shown promising performance on many
downstream visual tasks. With the surprising development of pre-trained visual foundation …

Pre-training by numerous image data has become de-facto for robust 2D representations. In
contrast, due to the expensive data processing, a paucity of 3D datasets severely hinders …

Recent advancements in multimodal pre-training have shown promising efficacy in 3D
representation learning by aligning multimodal features across 3D shapes their 2D …

Mainstream 3D representation learning approaches are built upon contrastive or generative
modeling pretext tasks, where great improvements in performance on various downstream …

This paper presents DreamLLM, a learning framework that first achieves versatile
Multimodal Large Language Models (MLLMs) empowered with frequently overlooked …

Large language models (LLMs) based on the generative pre-training transformer (GPT)
have demonstrated remarkable effectiveness across a diverse range of downstream tasks …

We introduce Point-Bind, a 3D multi-modality model aligning point clouds with 2D image,
language, audio, and video. Guided by ImageBind, we construct a joint embedding space …

Training a 3D scene understanding model requires complicated human annotations, which
are laborious to collect and result in a model only encoding close-set object semantics. In …

Self-supervised image backbones can be used to address complex 2D tasks (eg semantic
segmentation object discovery) very efficiently and with little or no downstream supervision …

The use of pretrained backbones with fine-tuning has been successful for 2D vision and
natural language processing tasks, showing advantages over task-specific networks. In this …