paper-with-me

Papers

PointAlign: Feature-Level Alignment Regularization for 3D Vision-Language Models

2026-02-28 · Yuanhao Su, Shaofeng Zhang, Xiaosong Jia, Qi Fan arxiv

The development of 3D Vision-Language Models (VLMs), crucial for applications in robotics, autonomous driving, and augmented reality, is severely constrained by the scarcity of paired 3D-text data. Existing methods rely solely on next-token prediction loss, using only language tokens for supervision. This results in inefficient utilization of limited 3D data and leads to a significant degradation and loss of valuable geometric information in intermediate representations. To address these limitations, we propose {\mname}, a novel feature-level alignment regularization method. {\mname} explicitly supervises intermediate point cloud tokens to preserve fine-grained 3D geometric-semantic information throughout the language modeling process. Specifically, we constrain the intermediate point cloud tokens within the LLM to align with visual input tokens via a consistency loss. By training only a lightweight alignment projector and LoRA adapters, {\mname} achieves explicit feature-level supervision with minimal computational overhead, effectively preventing geometric degradation. Extensive experiments on ModelNet40 and Objaverse datasets demonstrate that our method achieves \textbf{2.08} pp improvement on average for classification tasks, with a substantial \textbf{7.50} pp gain on the challenging open-vocabulary Objaverse classification task and \textbf{4.88} pp improvement on 3D object captioning evaluated by Qwen2-72B-Instruct, validating the effectiveness of {\mname}. Code is publicly available at \href{https://github.com/yharoldsu0627/PointAlign}{https://github.com/yharoldsu0627/PointAlign}.

📄 PDF Abstract BibTeX arXiv:2603.00412

Code (0)

등록된 구현이 없습니다.

Tasks

3D Object CaptioningAutonomous Driving

Similar Papers 제목 키워드 기반

Isotropic Embedding Perturbations for Robust Vision Language Encoders

2026-09-09 · Hyesong Choi, Daeun Kim, Song Park, Taekyung Kim 외 arxiv

Data augmentation is fundamental to training modern deep vision and multimodal models. While individual methods, such as RandAug, CutMix, Mixup, RandErase, and DropPath, offer strong regularization effects, their combine…

Data Augmentation

AIR-DA: Adversarial Image Reconstruction for Unsupervised Domain Adaptive Object Detection

2023-03-27 · Kunyang Sun, Wei Lin, Haoqin Shi, Zhengming Zhang 외

Unsupervised domain adaptive object detection is a challenging vision task where object detectors are adapted from a label-rich source domain to an unlabeled target domain. Recent advances prove the efficacy of the adver…

Image Reconstructionobject-detectionObject Detection

AttriPrompt: Dynamic Prompt Composition Learning for CLIP

2025-09-07 · Qiqi Zhan, Shiwei Li, Qingjie Liu, Yunhong Wang arxiv

The evolution of prompt learning methodologies has driven exploration of deeper prompt designs to enhance model performance. However, current deep text prompting approaches suffer from two critical limitations: Over-reli…

Contrastive Learning

Scalable Multi-Task Low-Rank Model Adaptation

2026-03-02 · Zichen Tian, Antoine Ledent, Qianru Sun arxiv

Scaling multi-task low-rank adaptation (LoRA) to a large number of tasks induces catastrophic performance degradation, such as an accuracy drop from 88.2% to 2.0% on DOTA when scaling from 5 to 15 tasks. This failure is …

Cross-modal Attention Congruence Regularization for Vision-Language Relation Alignment

2022-12-20 · Rohan Pandey, Rulin Shao, Paul Pu Liang, Ruslan Salakhutdinov 외

Despite recent progress towards scaling up multimodal vision-language models, these models are still known to struggle on compositional generalization benchmarks such as Winoground. We find that a critical component lack…

RelationVisual Reasoning