Image de couverture de livre

Efficient3D: A Unified Framework for Adaptive and Debiased Token Reduction in 3D MLLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Lin, Yuhui, Yu, Siyue, Yang, Yuxing, Cheng, Guangliang, Xiao, Jimin
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Hunyuan3D 1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation
par: Yang, Xianghui, et autres
Publié: (2024)

Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets
par: Hunyuan3D, Team, et autres
Publié: (2025)

Learning from Videos for 3D World: Enhancing MLLMs with 3D Vision Geometry Priors
par: Zheng, Duo, et autres
Publié: (2025)

TriCLIP-3D: A Unified Parameter-Efficient Framework for Tri-Modal 3D Visual Grounding based on CLIP
par: Li, Fan, et autres
Publié: (2025)

LTM3D: Bridging Token Spaces for Conditional 3D Generation with Auto-Regressive Diffusion Framework
par: Kang, Xin, et autres
Publié: (2025)

Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generation
par: Ye, Chongjie, et autres
Publié: (2026)

On the Limits of Token Reduction for Efficient Unified Vision Language Training
par: Chen, Siyi, et autres
Publié: (2026)

Artifact Reduction in Undersampled 3D Cone-Beam CTs using a Hybrid 2D-3D CNN Framework
par: Thalhammer, Johannes, et autres
Publié: (2026)

Multi-modal Relation Distillation for Unified 3D Representation Learning
par: Wang, Huiqun, et autres
Publié: (2024)

IDPruner: Harmonizing Importance and Diversity in Visual Token Pruning for MLLMs
par: Tan, Yifan, et autres
Publié: (2026)

Frequency-Aware Token Reduction for Efficient Vision Transformer
par: Lee, Dong-Jae, et autres
Publié: (2025)

SC3D: Label-Efficient Outdoor 3D Object Detection via Single Click Annotation
par: Xia, Qiming, et autres
Publié: (2024)

Wonder3D++: Cross-domain Diffusion for High-fidelity 3D Generation from a Single Image
par: Yang, Yuxiao, et autres
Publié: (2025)

SSGaussian: Semantic-Aware and Structure-Preserving 3D Style Transfer
par: Xu, Jimin, et autres
Publié: (2025)

Mamba-3D as Masked Autoencoders for Accurate and Data-Efficient Analysis of Medical Ultrasound Videos
par: Zhou, Jiaheng, et autres
Publié: (2025)

Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attention
par: Chen, Yiwen, et autres
Publié: (2025)

ForestFormer3D: A Unified Framework for End-to-End Segmentation of Forest LiDAR 3D Point Clouds
par: Xiang, Binbin, et autres
Publié: (2025)

Event USKT : U-State Space Model in Knowledge Transfer for Event Cameras
par: Lin, Yuhui, et autres
Publié: (2024)

VISA: Group-wise Visual Token Selection and Aggregation via Graph Summarization for Efficient MLLMs Inference
par: Jiang, Pengfei, et autres
Publié: (2025)

Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation
par: Mao, Jiawei, et autres
Publié: (2025)

Rethinking 3D Dense Caption and Visual Grounding in A Unified Framework through Prompt-based Localization
par: Luo, Yongdong, et autres
Publié: (2024)

MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models
par: Liu, Shengyuan, et autres
Publié: (2026)

S$^2$-MLLM: Boosting Spatial Reasoning Capability of MLLMs for 3D Visual Grounding with Structural Guidance
par: Xu, Beining, et autres
Publié: (2025)

Faster Parameter-Efficient Tuning with Token Redundancy Reduction
par: Kim, Kwonyoung, et autres
Publié: (2025)

Fre-Res: Frequency-Residual Video Token Compression for Efficient Video MLLMs
par: Feng, Yigui, et autres
Publié: (2026)

OccamToken: Efficient VLM Inference with Training-Free and Budget-Adaptive Token Pruning
par: Li, Geng, et autres
Publié: (2026)

InfoTok: Information-Theoretic Regularization for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs
par: Tang, Lv, et autres
Publié: (2026)

Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs
par: Zhang, Qizhe, et autres
Publié: (2025)

MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization
par: Li, Siyuan, et autres
Publié: (2025)

AdaTP: Attention-Debiased Token Pruning for Video Large Language Models
par: Sun, Fengyuan, et autres
Publié: (2025)

HY3D-Bench: Generation of 3D Assets
par: Hunyuan3D, Team, et autres
Publié: (2026)

2D-3D Interlaced Transformer for Point Cloud Segmentation with Scene-Level Supervision
par: Yang, Cheng-Kun, et autres
Publié: (2023)

From 2D Alignment to 3D Plausibility: Unifying Heterogeneous 2D Priors and Penetration-Free Diffusion for Occlusion-Robust Two-Hand Reconstruction
par: Han, Gaoge, et autres
Publié: (2025)

UniBEVFusion: Unified Radar-Vision BEVFusion for 3D Object Detection
par: Zhao, Haocheng, et autres
Publié: (2024)

Pandora3D: A Comprehensive Framework for High-Quality 3D Shape and Texture Generation
par: Yang, Jiayu, et autres
Publié: (2025)

Unifying 2D and 3D Vision-Language Understanding
par: Jain, Ayush, et autres
Publié: (2025)

Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation
par: Li, Wenhao, et autres
Publié: (2023)

UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encoding
par: Jiao, Yang, et autres
Publié: (2025)

Reinforced Embodied Active Defense: Exploiting Adaptive Interaction for Robust Visual Perception in Adversarial 3D Environments
par: Yang, Xiao, et autres
Publié: (2025)

UniLat3D: Geometry-Appearance Unified Latents for Single-Stage 3D Generation
par: Wu, Guanjun, et autres
Publié: (2025)