AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

LLaVA-3D: A Simple yet Effective Pathway to Empowering LMMs with 3D Capabilities

Chenming Zhu, Tai Wang, Wenwei Zhang, Jiangmiao Pang, Xihui Liu

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 197 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add 3D position encoding to 2D patches (3D patches) (ablation)LLaVA-3D · MMScan QAaccuracy: 42.1 -> 55.4 (+13.3)Improved
add 3D position encoding to 2D patches (3D patches) (ablation)LLaVA-3D · Scan2CapCIDEr: 29.7 -> 84.1 (+54.4)Improved
add 3D position encoding to 2D patches (ablation)LLaVA-3D · ScanQAmetric: 29.4 -> 29.8 (+0.4)No clear effect
add 3D position encoding to 2D patches (ablation)LLaVA-3D · SQA3Daccuracy: 59.8 -> 60.1 (+0.3)No clear effect
add grounding decoder instead of directly outputting text coordinates (ablation)LLaVA-3D · ScanReferAcc@0.25: 7.8 -> 50.1Improved
add grounding decoder instead of special-token coordinate output (ablation)LLaVA-3D · ScanReferAcc@0.25: 8.2 -> 50.1Improved
use CLIP only instead of SAM + CLIP for 3D scene features (ablation)3D LMM from Vicuna-7B · ScanRefer-related benchmarksinference time: 900s -> 0.2sImproved
use CLIP only instead of SAM + CLIP features (ablation)3D LMM from Vicuna-7B · ScanQAmetric: 22.1 -> 23.4Improved
use Pooling + MLP instead of Q-Former connector (ablation)3D LMM from Vicuna-7B · ScanQAmetric: 21.9 -> 22.1No clear effect
build from 2D LMM LLaVA-1.5 instead of LLM Vicuna (ablation)LLaVA-3D · ScanReferAcc@0.25: 43.8 -> 47.9Improved
use stronger 2D base (InternVL2.5-7B instead of LLaVA-1.5-7B) (ablation)LLaVA-3D · ScanQAmetric: 27.0 -> 29.1Improved
use video LMM base (LLaVA-Video-7B instead of InternVL2.5-7B) (ablation)LLaVA-3D · SQA3Daccuracy: 58.8 -> 60.1Improved

Later papers that compared with this method

1 later accepted paper reports LLaVA-3D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy