Move What Matters: Parameter-Efficient Domain Adaptation via Optimal Transport Flow for Collaborative Perception

2026-02-12

AI Summary

This work addresses the performance degradation and training instability of existing parameter-efficient fine-tuning methods in cross-domain deployment of multi-agent V2X cooperative perception, which stem from inter-frame redundancy and deep semantic degradation in heterogeneous perception streams. To tackle these challenges, we propose FlowAdapt, a novel framework that introduces optimal transport theory to this domain for the first time. FlowAdapt employs Wasserstein greedy sampling to eliminate redundant samples and incorporates a progressive knowledge transfer module that injects compressed shallow representations into deeper layers to mitigate semantic degradation. Requiring only 1% trainable parameters, our method significantly enhances cross-domain generalization and sample efficiency across three benchmarks, achieving an effective balance between semantic fidelity and computational efficiency.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Multiagent Systems: Adversarial Agents
• Computer Vision: Learning & Optimization for CV

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Fast domain adaptation remains a fundamental challenge for deploying multi-agent systems across diverse environments in Vehicle-to-Everything (V2X) collaborative perception. Despite the success of Parameter-Efficient Fine-Tuning (PEFT) in natural language processing and conventional vision tasks, directly applying PEFT to multi-agent settings leads to significant performance degradation and training instability. In this work, we conduct a detailed analysis and identify two key factors: (i) inter-frame redundancy in heterogeneous sensory streams, and (ii) erosion of fine-grained semantics in deep-layer representations under PEFT adaptation. To address these issues, we propose FlowAdapt, a parameter-efficient framework grounded in optimal transport theory, which minimizes information transport costs across both data distributions and network hierarchies. Specifically, we introduce a Wasserstein Greedy Sampling strategy to selectively filter redundant samples via a bounded covering radius. Furthermore, Progressive Knowledge Transfer module is designed to progressively inject compressed early-stage representations into later stages through learnable pathways, alleviating semantic degradation in late-stage adaptation. Extensive experiments on three benchmarks demonstrate that FlowAdapt achieves state-of-the-art performance with only 1% of trainable parameters, effectively bridging domain gaps with superior sample efficiency and generalization.

Problem

• domain adaptation
• collaborative perception
• parameter-efficient fine-tuning
• multi-agent systems
• semantic degradation

Innovation

• Parameter-Efficient Domain Adaptation
• Optimal Transport
• Collaborative Perception
• Wasserstein Sampling
• Progressive Knowledge Transfer