Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

EVA01: Unified Native 3D Understanding and Generation via Mixture-of-Transformers

About

This paper addresses the challenge of integrating 3D meshes as a native modality within Multimodal Large Language Models (MLLMs). Diffusion-based large reconstruction models decouple semantic understanding from geometric reasoning, operating as stateless reconstructors conditioned on dense 2D pixel priors. Recent MLLM-based methods treat the 3D modality as an external output rather than a native component of the multimodal sequence, making incremental adaptations without a systematic analysis of how geometric manifolds align with MLLM feature spaces. We introduce EVA01, a unified framework that extends the modality boundary of MLLMs to natively incorporate 3D mesh understanding, generation, and context-aware editing. Built upon a Mixture-of-Transformers (MoT) architecture, EVA01 decouples the model into a pre-trained Understanding Expert ($E_{\mathrm{und}}$) and a structurally mirrored Generation Expert ($E_{\mathrm{gen}}$), coupled through shared global self-attention with hard modality routing. This design aligns the semantic latent space of the MLLM backbone with the geometric manifold, enabling direct transfer of multimodal priors without intermediate 2D representations. Results show that EVA01 achieves state-of-the-art native text-to-3D generation fidelity and unlocks robust long-context multi-turn geometric editing with identity preservation, a capability fundamentally inaccessible to stateless reconstruction pipelines. Our findings further offer architectural insights for integrating 2D foundation models with 3D tasks, informing the design of 3D-native multimodal systems. Project Page: https://www.seeles.ai/research/pages/EVA01

Zongyuan Yang, Mingjing Yi, Wanli Ma, Chenzhuo Fan, Bocheng Li, Baolin Liu, Yuke Lou, Yingde Song, Yongping Xiong, Zhengdong Guo, Shimu Wang• 2026

Related benchmarks

TaskDatasetResultRank
Text-to-3DToys4k
CLIP Score35.72
25
3D Mesh CaptioningPointLLM-200 Objaverse
GPT Score (Image-based)65.91
11
Image-to-3DToys4k
CLIP Similarity87.28
11
Multi-turn 3D EditingPointLLM-200
CD0.018
4
Showing 4 of 4 rows

Other info

Follow for update