Jing Gao

Zeyu Zhou, Tianci Liu, Ruqi Bai et al.

In high-stake domains such as healthcare and hiring, the role of machine learning (ML) in decision-making raises significant fairness concerns. This work focuses on Counterfactual Fairness (CF), which posits that an ML model's outcome on any individual should remain unchanged if they had belonged to a different demographic group. Previous works have proposed methods that guarantee CF. Notwithstanding, their effects on the model's predictive performance remains largely unclear. To fill in this gap, we provide a theoretical study on the inherent trade-off between CF and predictive performance in a model-agnostic manner. We first propose a simple but effective method to cast an optimal but potentially unfair predictor into a fair one without losing the optimality. By analyzing its excess risk in order to achieve CF, we quantify this inherent trade-off. Further analysis on our method's performance with access to only incomplete causal knowledge is also conducted. Built upon it, we propose a performant algorithm that can be applied in such scenarios. Experiments on both synthetic and semi-synthetic datasets demonstrate the validity of our analysis and methods.

Zhuo Wang, Rongzhen Li, Bowen Dong et al.

Recent investigations show that large language models (LLMs), specifically GPT-4, not only have remarkable capabilities in common Natural Language Processing (NLP) tasks but also exhibit human-level performance on various professional and academic benchmarks. However, whether GPT-4 can be directly used in practical applications and replace traditional artificial intelligence (AI) tools in specialized domains requires further experimental validation. In this paper, we explore the potential of LLMs such as GPT-4 to outperform traditional AI tools in dementia diagnosis. Comprehensive comparisons between GPT-4 and traditional AI tools are conducted to examine their diagnostic accuracy in a clinical setting. Experimental results on two real clinical datasets show that, although LLMs like GPT-4 demonstrate potential for future advancements in dementia diagnosis, they currently do not surpass the performance of traditional AI tools. The interpretability and faithfulness of GPT-4 are also evaluated by comparison with real doctors. We discuss the limitations of GPT-4 in its current state and propose future research directions to enhance GPT-4 in dementia diagnosis.

Tianci Liu, Ruirui Li, Yunzhe Qi et al.

Large language models (LLMs) have achieved remarkable performance on various natural language tasks. However, they are trained on static corpora and their knowledge can become outdated quickly in the fast-changing world. This motivates the development of knowledge editing methods designed to update certain knowledge in LLMs without changing unrelated others. To make selective edits, previous efforts often sought to update a small amount of parameters in some specific layer(s) of a LLM. Nonetheless, in challenging scenarios, they still fall short in making successful edits while preserving knowledge irrelevant to the updates simultaneously, resulting in a notable editing-locality trade-off. In this work, we question if the trade-offs are caused by the fact that parameter-based updates have a global effect, i.e., edited parameters affect all inputs indiscriminately. In light of this, we explore the feasibility of representation fine-tuning, which applied some linear update to a few representations in a learned subspace, for knowledge editing. While being effective to enhance an LLM's general ability as demonstrated in the previous work, we theoretically show that this linear update imposes a tension in editing-locality trade-off. Subsequently, BaFT is proposed to break the linearity. BaFT computes a weight for each basis that spans a dimension of the subspace based on the input representation. This input-dependent weighting mechanism allows BaFT to manage different types of knowledge in an adaptive way, thereby achieving a better editing-locality trade-off. Experiments on three LLMs with five editing benchmarks in diverse scenarios show the superiority of our method.

Xingchen Wang, Feijie Wu, Chenglin Miao et al.

Split Federated Learning (SFL) has emerged as an efficient alternative to traditional Federated Learning (FL) by reducing client-side computation through model partitioning. However, exchanging of intermediate activations and model updates introduces significant privacy risks, especially from data reconstruction attacks that recover original inputs from intermediate representations. Existing defenses using noise injection often degrade model performance. To overcome these challenges, we present PM-SFL, a scalable and privacy-preserving SFL framework that incorporates Probabilistic Mask training to add structured randomness without relying on explicit noise. This mitigates data reconstruction risks while maintaining model utility. To address data heterogeneity, PM-SFL employs personalized mask learning that tailors submodel structures to each client's local data. For system heterogeneity, we introduce a layer-wise knowledge compensation mechanism, enabling clients with varying resources to participate effectively under adaptive model splitting. Theoretical analysis confirms its privacy protection, and experiments on image and wireless sensing tasks demonstrate that PM-SFL consistently improves accuracy, communication efficiency, and robustness to privacy attacks, with particularly strong performance under data and system heterogeneity.

Haoyu Liu, Fenglong Ma, Shibo He et al.

Outlier ensemble methods have shown outstanding performance on the discovery of instances that are significantly different from the majority of the data. However, without the awareness of fairness, their applicability in the ethical scenarios, such as fraud detection and judiciary judgement system, could be degraded. In this paper, we propose to reduce the bias of the outlier ensemble results through a fairness-aware ensemble framework. Due to the lack of ground truth in the outlier detection task, the key challenge is how to mitigate the degradation in the detection performance with the improvement of fairness. To address this challenge, we define a distance measure based on the output of conventional outlier ensemble techniques to estimate the possible cost associated with detection performance degradation. Meanwhile, we propose a post-processing framework to tune the original ensemble results through a stacking process so that we can achieve a trade off between fairness and detection performance. Detection performance is measured by the area under ROC curve (AUC) while fairness is measured at both group and individual level. Experiments on eight public datasets are conducted. Results demonstrate the effectiveness of the proposed framework in improving fairness of outlier ensemble results. We also analyze the trade-off between AUC and fairness.

Yaqing Wang, Fenglong Ma, Jing Gao

Recent advances in information extraction have motivated the automatic construction of huge Knowledge Graphs (KGs) by mining from large-scale text corpus. However, noisy facts are unavoidably introduced into KGs that could be caused by automatic extraction. To validate the correctness of facts (i.e., triplets) inside a KG, one possible approach is to map the triplets into vector representations by capturing the semantic meanings of facts. Although many representation learning approaches have been developed for knowledge graphs, these methods are not effective for validation. They usually assume that facts are correct, and thus may overfit noisy facts and fail to detect such facts. Towards effective KG validation, we propose to leverage an external human-curated KG as auxiliary information source to help detect the errors in a target KG. The external KG is built upon human-curated knowledge repositories and tends to have high precision. On the other hand, although the target KG built by information extraction from texts has low precision, it can cover new or domain-specific facts that are not in any human-curated repositories. To tackle this challenging task, we propose a cross-graph representation learning framework, i.e., CrossVal, which can leverage an external KG to validate the facts in the target KG efficiently. This is achieved by embedding triplets based on their semantic meanings, drawing cross-KG negative samples and estimating a confidence score for each triplet based on its degree of correctness. We evaluate the proposed framework on datasets across different domains. Experimental results show that the proposed framework achieves the best performance compared with the state-of-the-art methods on large-scale KGs.

Hengtong Zhang, Tianhang Zheng, Jing Gao et al.

Knowledge graph embedding (KGE) is a technique for learning continuous embeddings for entities and relations in the knowledge graph.Due to its benefit to a variety of downstream tasks such as knowledge graph completion, question answering and recommendation, KGE has gained significant attention recently. Despite its effectiveness in a benign environment, KGE' robustness to adversarial attacks is not well-studied. Existing attack methods on graph data cannot be directly applied to attack the embeddings of knowledge graph due to its heterogeneity. To fill this gap, we propose a collection of data poisoning attack strategies, which can effectively manipulate the plausibility of arbitrary targeted facts in a knowledge graph by adding or deleting facts on the graph. The effectiveness and efficiency of the proposed attack strategies are verified by extensive evaluations on two widely-used benchmarks.

Chenwei Zhang, Sihong Xie, Yaliang Li et al.

In big data applications such as healthcare data mining, due to privacy concerns, it is necessary to collect predictions from multiple information sources for the same instance, with raw features being discarded or withheld when aggregating multiple predictions. Besides, crowd-sourced labels need to be aggregated to estimate the ground truth of the data. Because of the imperfect predictive models or human crowdsourcing workers, noisy and conflicting information is ubiquitous and inevitable. Although state-of-the-art aggregation methods have been proposed to handle label spaces with flat structures, as the label space is becoming more and more complicated, aggregation under a label hierarchical structure becomes necessary but has been largely ignored. These label hierarchies can be quite informative as they are usually created by domain experts to make sense of highly complex label correlations for many real-world cases like protein functionality interactions or disease relationships. We propose a novel multi-source hierarchical prediction consolidation method to effectively exploits the complicated hierarchical label structures to resolve the noisy and conflicting information that inherently originates from multiple imperfect sources. We formulate the problem as an optimization problem with a closed-form solution. The proposed method captures the smoothness overall information sources as well as penalizing any consolidation result that violates the constraints derived from the label hierarchy. The hierarchical instance similarity, as well as the consolidation result, are inferred in a totally unsupervised, iterative fashion. Experimental results on both synthetic and real-world datasets show the effectiveness of the proposed method over existing alternatives.