Author name cluster

Zhen Sun

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

9 papers

2 author rows

AAAI Conference 2026 Conference Paper

6DAttack: Backdoor Attacks in the 6DoF Pose Estimation

Jihui Guo
Zongmin Zhang
Zhen Sun
Yuhao Yang
Jinlin Wu
Fu Zhang
Xinlei He

Recent advances in deep learning have enabled highly accurate six-degree-of-freedom (6DoF) object pose estimation, leading to its widespread use in real-world applications such as robotics, augmented reality, virtual reality, and autonomous systems. However, backdoor attacks pose a major security risk to deep learning models. By injecting malicious triggers into training data, an attacker can cause a model to perform normally on benign inputs but behave incorrectly under specific conditions. While most research on backdoor attacks has focused on 2D vision tasks, their impact on 6DoF pose estimation remains largely unexplored. Furthermore, unlike traditional backdoors that only change the object class, backdoors against 6DoF pose estimation must additionally control continuous pose parameters, such as translation and rotation, making existing 2D backdoor attack methods not directly applicable to this setting. To address this gap, we propose a novel backdoor attack framework (6DAttack) that exposes vulnerabilities in 6DoF pose estimation. 6DAttack uses synthetic and real 3D objects of varying shapes as triggers and assigns target poses to induce controlled erroneous pose outputs while maintaining normal behavior on clean inputs. We evaluated this attack on multiple models (including PVNet, DenseFusion, and PoseDiffusion) and datasets (including LINEMOD, YCB-Video, and CO3D). Experimental results demonstrate that 6DAttack achieves extremely high attack success rates (ASRs) without compromising performance on legitimate tasks. Across various models and objects, the backdoored models achieve up to 100% ADD accuracy on clean data, while also achieving 100% ASR under trigger conditions. The accuracy of controlled erroneous pose output is also extremely high, with triggered samples achieving 97.70% ADD-P. These results demonstrate that the backdoor can be reliably implanted and activated, achieving a high ASR under trigger conditions while maintaining a negligible impact on benign data. Furthermore, we evaluate a representative defense and show that it remains ineffective under 6DAttack. Overall, our findings reveal a potentially serious and previously underexplored threat to modern 6DoF pose estimation models.

PDF Details DOI

IROS Conference 2025 Conference Paper

A 2 I-Calib: An Anti-Noise Active Multi-IMU Spatial-Temporal Calibration Framework for Legged Robots

Chaoran Xiong
Fangyu Jiang
Kehui Ma
Zhen Sun
Zeyu Zhang
Ling Pei

Recently, multi-node inertial measurement unit (IMU)-based odometry for legged robots has gained attention due to its cost-effectiveness, power efficiency, and high accuracy. However, the spatial and temporal misalignment between foot-end motion derived from forward kinematics and foot IMU measurements can introduce inconsistent constraints, resulting in odometry drift. Therefore, accurate spatial-temporal calibration is crucial for the multi-IMU systems. Although existing multi-IMU calibration methods have addressed passive single-rigid-body sensor calibration, they are inadequate for legged systems. This is due to the insufficient excitation from traditional gaits for calibration, and enlarged sensitivity to IMU noise during kinematic chain transformations. To address these challenges, we propose A 2 I-Calib, an anti-noise active multi-IMU calibration framework enabling autonomous spatial-temporal calibration for arbitrary foot-mounted IMUs. Our A 2 I-Calib includes: 1) an anti-noise trajectory generator leveraging a proposed basis function selection theorem to minimize the condition number in correlation analysis, thus reducing noise sensitivity, and 2) a reinforcement learning (RL)-based controller that ensures robust execution of calibration motions. Furthermore, A 2 I-Calib is validated on simulation and real-world quadruped robot platforms with various multi-IMU settings, which demonstrates a significant reduction in noise sensitivity and calibration errors, thereby improving the overall multi-IMU odometry performance.

Details

NeurIPS Conference 2025 Conference Paper

CHASM: Unveiling Covert Advertisements on Chinese Social Media

Jingyi Zheng
Tianyi Hu
Yule Liu
Zhen Sun
Zongmin Zhang
Zifan Peng
Wenhan Dong
Xinlei He

Current benchmarks for evaluating large language models (LLMs) in social media moderation completely overlook a serious threat: covert advertisements, which disguise themselves as regular posts to deceive and mislead consumers into making purchases, leading to significant ethical and legal concerns. In this paper, we present the CHASM, a first-of-its-kind dataset designed to evaluate the capability of Multimodal Large Language Models (MLLMs) in detecting covert advertisements on social media. CHASM is a high-quality, anonymized, manually curated dataset consisting of 4, 992 instances, based on real-world scenarios from the Chinese social media platform Rednote. The dataset was collected and annotated under strict privacy protection and quality control protocols. It includes many product experience sharing posts that closely resemble covert advertisements, making the dataset particularly challenging. The results show that under both zero-shot and in-context learning settings, none of the current MLLMs are sufficiently reliable for detecting covert advertisements. Our further experiments revealed that fine-tuning open-source MLLMs on our dataset yielded noticeable performance gains. However, significant challenges persist, such as detecting subtle cues in comments and differences in visual and textual structures. We provide in-depth error analysis and outline future research directions. We hope our study can serve as a call for the research community and platform moderators to develop more precise defenses against this emerging threat.

PDF Details

ICML Conference 2025 Conference Paper

DS-VLM: Diffusion Supervision Vision Language Model

Zhen Sun
Yunhang Shen
Jie Li 0052
Xing Sun 0001
Pingyang Dai
Liujuan Cao
Rongrong Ji

Vision-Language Models (VLMs) face two critical limitations in visual representation learning: degraded supervision due to information loss during gradient propagation, and the inherent semantic sparsity of textual supervision compared to visual data. We propose the Diffusion Supervision Vision-Language Model (DS-VLM), a plug-and-play framework that introduces diffusion-based direct supervision for vision-language alignment. By reconstructing input images through a diffusion model conditioned on outputs of the visual encoder and the connector, our method establishes a short-path gradient propagation channel from pixel space to visual features. This approach simultaneously preserves high-level semantic alignment through conventional text supervision while enhancing visual feature quality via pixel-level reconstruction constraints. Extensive experiments conducted across various visual encoders and LLMs of different scales demonstrate the effectiveness of our approach.

Details

ICML Conference 2025 Conference Paper

FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification

Zhen Sun
Lei Tan
Yunhang Shen
Chengmao Cai
Xing Sun 0001
Pingyang Dai
Liujuan Cao
Rongrong Ji

Multimodal person re-identification (Re-ID) aims to match pedestrian images across different modalities. However, most existing methods focus on limited cross-modal settings and fail to support arbitrary query-retrieval combinations, hindering practical deployment. We propose FlexiReID, a flexible framework that supports seven retrieval modes across four modalities: RGB, infrared, sketches, and text. FlexiReID introduces an adaptive mixture-of-experts (MoE) mechanism to dynamically integrate diverse modality features and a cross-modal query fusion module to enhance multimodal feature extraction. To facilitate comprehensive evaluation, we construct CIRS-PEDES, a unified dataset extending four popular Re-ID datasets to include all four modalities. Extensive experiments demonstrate that FlexiReID achieves state-of-the-art performance and offers strong generalization in complex scenarios.

Details

ICRA Conference 2025 Conference Paper

IMOST: Incremental Memory Mechanism with Online Self-Supervision for Continual Traversability Learning

Kehui Ma
Zhen Sun
Chaoran Xiong
Qiumin Zhu
Kewei Wang 0009
Ling Pei

Traversability estimation is the foundation of path planning for a general navigation system. However, complex and dynamic environments pose challenges for the latest methods using self-supervised learning (SSL) technique. Firstly, existing SSL-based methods generate sparse annotations lacking detailed boundary information. Secondly, their strategies focus on hard samples for rapid adaptation, leading to forgetting and biased predictions. In this work, we propose IMOST, a continual traversability learning framework composed of two key modules: incremental dynamic memory (IDM) and self-supervised annotation (SSA). By mimicking human memory mechanisms, IDM allocates novel data samples to new clusters according to information expansion criterion. It also updates clusters based on diversity rule, ensuring a representative characterization of new scene. This mechanism enhances scene-aware knowledge diversity while maintaining a compact memory capacity. The SSA module, integrating FastSAM, utilizes point prompts to generate complete annotations in real time which reduces training complexity. Furthermore, IMOST has been successfully deployed on the quadruped robot, with performance evaluated during the online learning process. Experimental results on both public and self-collected datasets demonstrate that our IMOST outperforms current state-of-the-art method, maintains robust recognition capabilities and adaptability across various scenarios. The code is available at https://github.com/SJTUMKH/OCLTrav.

Details

EAAI Journal 2025 Journal Article

Seismic damage states prediction of in-service bridges using feature-enhanced swin transformer without reliance on damage indicators

Yalin Li
Zhen Sun
Sujith Mangalathu
Yaqi Li
Hao Yang
Weidong He

To achieve real-time and efficient evaluation of seismic damage to structures, this study proposes an improved deep learning-based model, the deep feature-enhanced Swin Transformer (CC-SwinT). This model overcomes the influence of service life on the time-varying damage indicators of bridges. By eliminating the reliance on seismic performance and damage indicators, it predicts the seismic damage state of in-service bridges based solely on the response of structure. The CC-SwinT model integrates continuous wavelet transform (CWT) technology and the context anchored attention (CAA) mechanism to enhance the extraction of structure response features of bridge piers. This integration enables the model to effectively mine time-frequency characteristics and capture non-local long-term dependencies in structure responses. To comprehensively train the CC-SwinT model, a structure response database for in-service bridges was constructed based on a data-driven objectives, analyzing the impacts of service conditions on the seismic performance of bridges. Subsequently, transfer learning methods were applied, and the performance of the CC-SwinT framework was evaluated using various metrics to highlight its exceptional feature extraction and prediction capabilities. Furthermore, the Gradient-weighted Class Activation Mapping (Grad-CAM) interpretability technique was used to explore the decision-making process and feature focus of CC-SwinT. The findings of this study provide a valuable reference for seismic damage prediction of in-service structures and rapid post-earthquake rescue response.

Details DOI

IROS Conference 2025 Conference Paper

THE-SEAN: A Heart Rate Variation-Inspired Temporally High-Order Event-Based Visual Odometry with Self-Supervised Spiking Event Accumulation Networks

Chaoran Xiong
Litao Wei
Kehui Ma
Zhen Sun
Yan Xiang
Zihan Nan
Trieu-Kien Truong
Ling Pei

Event-based visual odometry has recently gained attention for its high accuracy and real-time performance in fast-motion systems. Unlike traditional synchronous estimators that rely on constant-frequency (zero-order) triggers, event-based visual odometry can actively accumulate information to generate temporally high-order estimation triggers. However, existing methods primarily focus on adaptive event representation after estimation triggers, neglecting the decision-making process for efficient temporal triggering itself. This oversight leads to the computational redundancy and noise accumulation. In this paper, we introduce a temporally high-order event-based visual odometry with spiking event accumulation networks (THE-SEAN). To the best of our knowledge, it is the first event-based visual odometry capable of dynamically adjusting its estimation trigger decision in response to motion and environmental changes. Inspired by biological systems that regulate hormone secretion to modulate heart rate, a self-supervised spiking neural network is designed to generate estimation triggers. This spiking network extracts temporal features to produce triggers, with rewards based on block matching points and Fisher information matrix (FIM) trace acquired from the estimator itself. Finally, THE-SEAN is evaluated across several open datasets, thereby demonstrating average improvements of 13% in estimation accuracy, 9% in smoothness, and 38% in triggering efficiency compared to the state-of-the-art methods.

Details

NeurIPS Conference 2020 Conference Paper

De-Anonymizing Text by Fingerprinting Language Generation

Zhen Sun
Roei Schuster
Vitaly Shmatikov

Components of machine learning systems are not (yet) perceived as security hotspots. Secure coding practices, such as ensuring that no execution paths depend on confidential inputs, have not yet been adopted by ML developers. We initiate the study of code security of ML systems by investigating how nucleus sampling---a popular approach for generating text, used for applications such as auto-completion---unwittingly leaks texts typed by users. Our main result is that the series of nucleus sizes for many natural English word sequences is a unique fingerprint. We then show how an attacker can infer typed text by measuring these fingerprints via a suitable side channel (e. g. , cache access times), explain how this attack could help de-anonymize anonymous texts, and discuss defenses.

PDF Details