Cover Image

Does Your 3D Encoder Really Work? When Pretrain-SFT from 2D VLMs Meets 3D VLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Haoyuan, Zhou, Yanpeng, Gao, Yufei, Tang, Tao, Han, Jianhua, Yuan, Yujie, Chen, Dave Zhenyu, Bian, Jiawang, Xu, Hang, Liang, Xiaodan
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

UniGS: Unified Language-Image-3D Pretraining with Gaussian Splatting
by: Li, Haoyuan, et al.
Published: (2025)

GS-CLIP: Gaussian Splatting for Contrastive Language-Image-3D Pretraining from Real-World Data
by: Li, Haoyuan, et al.
Published: (2024)

When Does RL Help Medical VLMs? Disentangling Vision, SFT, and RL Gains
by: Jeddi, Ahmadreza, et al.
Published: (2026)

MetaSpatial: Reinforcing 3D Spatial Reasoning in VLMs for the Metaverse
by: Pan, Zhenyu, et al.
Published: (2025)

Are VLMs Really Blind
by: Singh, Ayush, et al.
Published: (2024)

VaseVQA-3D: Benchmarking 3D VLMs on Ancient Greek Pottery
by: Zhang, Nonghai, et al.
Published: (2025)

Agentic 3D Scene Generation with Spatially Contextualized VLMs
by: Liu, Xinhang, et al.
Published: (2025)

3D-R1: Enhancing Reasoning in 3D VLMs for Unified Scene Understanding
by: Huang, Ting, et al.
Published: (2025)

3D Primitives are a Spatial Language for VLMs
by: Liu, Junze, et al.
Published: (2026)

Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning
by: He, Jixuan, et al.
Published: (2026)

When Data Manipulation Meets Attack Goals: An In-depth Survey of Attacks for VLMs
by: Dai, Aobotao, et al.
Published: (2025)

The Synergy Dilemma of Long-CoT SFT and RL: Investigating Post-Training Techniques for Reasoning VLMs
by: Chen, Jierun, et al.
Published: (2025)

Monocular 3D Object Position Estimation with VLMs for Human-Robot Interaction
by: Wahl, Ari, et al.
Published: (2026)

Global-Local Tree Search in VLMs for 3D Indoor Scene Generation
by: Deng, Wei, et al.
Published: (2025)

D-CoDe: Scaling Image-Pretrained VLMs to Video via Dynamic Compression and Question Decomposition
by: Huang, Yiyang, et al.
Published: (2025)

DreamVTON: Customizing 3D Virtual Try-on with Personalized Diffusion Models
by: Xie, Zhenyu, et al.
Published: (2024)

MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders
by: Cao, Jiajun, et al.
Published: (2025)

Have the VLMs Lost Confidence? A Study of Sycophancy in VLMs
by: Li, Shuo, et al.
Published: (2024)

VisualOverload: Probing Visual Understanding of VLMs in Really Dense Scenes
by: Gavrikov, Paul, et al.
Published: (2025)

Reinforcing 3D Understanding in Point-VLMs via Geometric Reward Credit Assignment
by: Chen, Jingkun, et al.
Published: (2026)

LLaVA$^3$: Representing 3D Scenes like a Cubist Painter to Boost 3D Scene Understanding of VLMs
by: Petit, Doriand, et al.
Published: (2025)

When VLMs Meet Image Classification: Test Sets Renovation via Missing Label Identification
by: Pang, Zirui, et al.
Published: (2025)

From Thousands to Billions: 3D Visual Language Grounding via Render-Supervised Distillation from 2D VLMs
by: Cao, Ang, et al.
Published: (2025)

BabyVLM: Data-Efficient Pretraining of VLMs Inspired by Infant Learning
by: Wang, Shengao, et al.
Published: (2025)

Does the Question Really Matter? Training-Free Data Selection for Vision-Language SFT
by: Sun, Peng, et al.
Published: (2026)

Collaboration: It Really Does Work!
by: Youssef, Jennifer L.
Published: (2005)

Drive-P2D: A Progressive Perception-to-Decision Benchmark for VLMs in Autonomous Driving
by: Tang, Zecong, et al.
Published: (2026)

Level Up Your Tutorials: VLMs for Game Tutorials Quality Assessment
by: Cambrin, Daniele Rege, et al.
Published: (2024)

Should VLMs be Pre-trained with Image Data?
by: Keh, Sedrick, et al.
Published: (2025)

How Do VLAs Effectively Inherit from VLMs?
by: Zhang, Chuheng, et al.
Published: (2025)

Thinking with Geometry: Active Geometry Integration for Spatial Reasoning
by: Li, Haoyuan, et al.
Published: (2026)

Do 3D Large Language Models Really Understand 3D Spatial Relationships?
by: Ma, Xianzheng, et al.
Published: (2026)

Do VLMs Need Vision Transformers? Evaluating State Space Models as Vision Encoders
by: Kuo, Shang-Jui Ray, et al.
Published: (2026)

VLMs have Tunnel Vision: Evaluating Nonlocal Visual Reasoning in Leading VLMs
by: Berman, Shmuel, et al.
Published: (2025)

3D Feature Prediction for Masked-AutoEncoder-Based Point Cloud Pretraining
by: Yan, Siming, et al.
Published: (2023)

AnchorSplat: Feed-Forward 3D Gaussian Splatting with 3D Geometric Priors
by: Zhang, Xiaoxue, et al.
Published: (2026)

3D-GSRD: 3D Molecular Graph Auto-Encoder with Selective Re-mask Decoding
by: Wu, Chang, et al.
Published: (2025)

Hyper3D: Efficient 3D Representation via Hybrid Triplane and Octree Feature for Enhanced 3D Shape Variational Auto-Encoders
by: Guo, Jingyu, et al.
Published: (2025)

Bringing Your Portrait to 3D Presence
by: Zhang, Jiawei, et al.
Published: (2025)

DEAL: Disentangle and Localize Concept-level Explanations for VLMs
by: Li, Tang, et al.
Published: (2024)