# CP-Guard+: A New Paradigm for Malicious Agent Detection and Defense in Collaborative Perception

arXiv: https://arxiv.org/abs/2502.07807 · HTML: https://mlsanity.com/paper/2502.07807/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Feb 2025.

## Built on by (most cited first, 7 of 7)

- [Distribution-Aligned Decoding for Efficient LLM Task Adaptation](https://mlsanity.com/paper/2509.15888/) (2025)
- [Task-Aware Parameter-Efficient Fine-Tuning of Large Pre-Trained Models at the Edge](https://mlsanity.com/paper/2504.03718/) (2025)
- [CP-uniGuard: A Unified, Probability-Agnostic, and Adaptive Framework for Malicious Agent Detection and Defense in Multi-Agent Embodied Perception Systems](https://mlsanity.com/paper/2506.22890/) (2025)
- [All Vehicles Can Lie: Efficient Adversarial Defense in Fully Untrusted-Vehicle Collaborative Perception via Pseudo-Random Bayesian Inference](https://mlsanity.com/paper/2603.08498/) (2026)
- [HyDRA: Hybrid Domain-Aware Robust Architecture for Heterogeneous Collaborative Perception](https://mlsanity.com/paper/2603.23975/) (2026)
- [Adversarial Trust Poisoning in Vehicular Collaborative Perception](https://mlsanity.com/paper/2605.22122/) (2026)
- [Multi-Agent Embodied Autonomous Driving (MAEAD): From V2X Information Exchange to Shared World Models](https://mlsanity.com/paper/2606.13840/) (2026)

## Abstract (arXiv)

Collaborative perception (CP) is a promising method for safe connected and autonomous driving, which enables multiple vehicles to share sensing information to enhance perception performance. However, compared with single-vehicle perception, the openness of a CP system makes it more vulnerable to malicious attacks that can inject malicious information to mislead the perception of an ego vehicle, resulting in severe risks for safe driving. To mitigate such vulnerability, we first propose a new paradigm for malicious agent detection that effectively identifies malicious agents at the feature level without requiring verification of final perception results, significantly reducing computational overhead. Building on this paradigm, we introduce CP-GuardBench, the first comprehensive dataset provided to train and evaluate various malicious agent detection methods for CP systems. Furthermore, we develop a robust defense method called CP-Guard+, which enhances the margin between the representations of benign and malicious features through a carefully designed Dual-Centered Contrastive Loss (DCCLoss). Finally, we conduct extensive experiments on both CP-GuardBench and V2X-Sim, and demonstrate the superiority of CP-Guard+.