Author name cluster

Mohan Li

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.

5 papers

1 author row

AAAI Conference 2026 Conference Paper

AT-Field: Rethinking the Games in Adversarial Training

Yixiao Xu
Mohan Li
Zhijie Shen
Yuan Liu
Zhihong Tian

Adversarial training is often modeled as a two-player zero-sum game, relying on strong assumptions that limit its practical guidance. In this paper, we instead analyze the interactions between training samples and show that even the fundamental objective—minimizing training loss—may not converge. To address this, we propose AT-Field, an adversarial training framework guided by sample-wise game-theoretic relationships. Specifically, we prove that training samples across different batches can form a none-potential game, where gradient descent induces cyclic behaviors, preventing convergence. By strategically searching and grouping these samples within the same batch, AT-Field transforms none-potential games into exact potential games, which are more effectively optimized using gradient-based methods. Experiments demonstrate that AT-Field integrates seamlessly with existing adversarial training techniques, enhancing both accuracy and robustness.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

FLUX: Efficient Descriptor-Driven Clustered Federated Learning under Arbitrary Distribution Shifts

Dario Fenoglio
Mohan Li
Pietro Barbiero
Nicholas Lane
Marc Langheinrich
Martin Gjoreski

Federated Learning (FL) enables collaborative model training across multiple clients while preserving data privacy. Traditional FL methods often use a global model to fit all clients, assuming that clients' data are independent and identically distributed (IID). However, when this assumption does not hold, the global model accuracy may drop significantly, limiting FL applicability in real-world scenarios. To address this gap, we propose FLUX, a novel clustering-based FL (CFL) framework that addresses the four most common types of distribution shifts during both training and test time. To this end, FLUX leverages privacy-preserving client-side descriptor extraction and unsupervised clustering to ensure robust performance and scalability across varying levels and types of distribution shifts. Unlike existing CFL methods addressing non-IID client distribution shifts, FLUX i) does not require any prior knowledge of the types of distribution shifts or the number of client clusters, and ii) supports test-time adaptation, enabling unseen and unlabeled clients to benefit from the most suitable cluster-specific models. Extensive experiments across four standard benchmarks, two real-world datasets and ten state-of-the-art baselines show that FLUX improves performance and stability under diverse distribution shifts—achieving an average accuracy gain of up to 23 percentage points over the best-performing baselines—while maintaining computational and communication overhead comparable to FedAvg.

PDF Details

AAAI Conference 2025 Conference Paper

Mitigating Hallucinations in Large Vision-Language Models by Adaptively Constraining Information Flow

Jiaqi Bai
Hongcheng Guo
Zhongyuan Peng
Jian Yang
Zhoujun Li
Mohan Li
Zhihong Tian

Large vision-language models show tremendous potential in understanding visual information through human languages. However, they are prone to suffer from object hallucination, i.e., the generated image descriptions contain objects that do not exist in the image. In this paper, we reveal that object hallucination can be attributed to overconfidence in irrelevant visual features when soft visual tokens map to the LLM's word embedding space. Specifically, by figuring out the semantic similarity between visual tokens and LLM's word embedding, we observe that the smoothness of similarity distribution strongly correlates with the emergence of object hallucinations. To mitigate hallucinations, we propose using the Variational Information Bottleneck (VIB) to alleviate overconfidence by introducing stochastic noise, facilitating the constraining of irrelevant information. Furthermore, we propose an entropy-based noise-controlling strategy to enable the injected noise to be adaptively constrained regarding the smoothness of the similarity distribution. We adapt the proposed AdaVIB across distinct model architectures. Experimental results demonstrate that the proposed AdaVIB mitigates object hallucinations by effectively alleviating the overconfidence in irrelevant visual features, with consistent improvements on two object hallucination benchmarks.

PDF Details DOI

JBHI Journal 2025 Journal Article

Virtual-mask Informed Prior for Sparse-view Dual-Energy CT Reconstruction

Zini Chen
Yao Xiao
Junyan Zhang
Mohan Li
Cunfeng Wei
Shaoyu Wang
Liu Shi
Qiegen Liu

Sparse-view sampling in dual-energy computed tomography (DECT) significantly reduces radiation dose and increases imaging speed, yet is highly prone to artifacts. Although diffusion models have demonstrated potential in effectively handling incomplete data, most existing methods in this field focus on the image domain and lack global constraints, which consequently leads to insufficient reconstruction quality. In this study, we propose a dual-domain virtual-mask informed diffusion model for sparse-view reconstruction by leveraging the high inter-channel correlation in DECT. Specifically, the study designs a virtual mask and applies it to the high-energy and low-energy data to perform perturbation operations, thus constructing high-dimensional tensors that serve as the prior information of the diffusion model. In addition, a dual-domain collaboration strategy is adopted to integrate the information of the randomly selected high-frequency components in the wavelet domain with the information in the projection domain, for the purpose of optimizing the global structures and local details. The experimental results show that the method exhibits excellent performance on multiple datasets. Under 30-view sparse sampling conditions, VIP-DECT improves PSNR by at least 1. 02 dB and enhances SSIM by 1. 91%.

Details DOI

NeurIPS Conference 2024 Conference Paper

LT-Defense: Searching-free Backdoor Defense via Exploiting the Long-tailed Effect

Yixiao Xu
Binxing Fang
Mohan Li
Keke Tang
Zhihong Tian

Language models have shown vulnerability against backdoor attacks, threatening the security of services based on them. To mitigate the threat, existing solutions attempted to search for backdoor triggers, which can be time-consuming when handling a large search space. Looking into the attack process, we observe that poisoned data will create a long-tailed effect in the victim model, causing the decision boundary to shift towards the attack targets. Inspired by this observation, we introduce LT-Defense, the first searching-free backdoor defense via exploiting the long-tailed effect. Specifically, LT-Defense employs a small set of clean examples and two metrics to distinguish backdoor-related features in the target model. Upon detecting a backdoor model, LT-Defense additionally provides test-time backdoor freezing and attack target prediction. Extensive experiments demonstrate the effectiveness of LT-Defense in both detection accuracy and efficiency, e. g. , in task-agnostic scenarios, LT-Defense achieves 98% accuracy across 1440 models with less than 1% of the time cost of state-of-the-art solutions.

PDF Details DOI