Mantis: Mamba-native Tuning is Efficient for 3D Point Cloud Foundation Models

TL;DR

Mantis introduces a Mamba-native PEFT framework for 3D point cloud models, enabling efficient fine-tuning with minimal parameters while maintaining high accuracy and stability.

Contribution

The paper presents Mantis, the first PEFT method tailored for Mamba-based 3D PFMs, incorporating a State-Aware Adapter and Dual-Serialization Consistency Distillation.

Findings

Achieves competitive performance with only 5% trainable parameters.

Addresses instability caused by point cloud serialization.

Demonstrates effectiveness across multiple benchmarks.

Abstract

Pre-trained 3D point cloud foundation models (PFMs) have demonstrated strong transferability across diverse downstream tasks. However, full fine-tuning these models is computationally expensive and storage-intensive. Parameter-efficient fine-tuning (PEFT) offers a promising alternative, but existing PEFT approaches are primarily designed for Transformer-based backbones and rely on token-level prompting or feature transformation. Mamba-based backbones introduce a granularity mismatch between token-level adaptation and state-level sequence dynamics. Consequently, straightforward transfer of existing PEFT approaches to frozen Mamba backbones leads to substantial accuracy degradation and unstable optimization. To address this issue, we propose Mantis, the first Mamba-native PEFT framework for 3D PFMs. Specifically, a State-Aware Adapter (SAA) is introduced to inject lightweight task-conditioned control signals into selective state-space updates, enabling state-level adaptation while keeping the pre-trained backbone frozen. Moreover, different valid point cloud serializations are regularized by Dual-Serialization Consistency Distillation (DSCD), thereby reducing serialization-induced instability. Extensive experiments across multiple benchmarks demonstrate that our Mantis achieves competitive performance with only about 5% trainable parameters. Our code is available at https://github.com/gzhhhhhhh/Mantis.