Jiankai Sun

Zhaowei Zhu, Yuanshun Yao, Jiankai Sun et al.

Evaluating fairness can be challenging in practice because the sensitive attributes of data are often inaccessible due to privacy constraints. The go-to approach that the industry frequently adopts is using off-the-shelf proxy models to predict the missing sensitive attributes, e.g. Meta [Alao et al., 2021] and Twitter [Belli et al., 2022]. Despite its popularity, there are three important questions unanswered: (1) Is directly using proxies efficacious in measuring fairness? (2) If not, is it possible to accurately evaluate fairness using proxies only? (3) Given the ethical controversy over inferring user private information, is it possible to only use weak (i.e. inaccurate) proxies in order to protect privacy? Our theoretical analyses show that directly using proxy models can give a false sense of (un)fairness. Second, we develop an algorithm that is able to measure fairness (provably) accurately with only three properly identified proxies. Third, we show that our algorithm allows the use of only weak proxies (e.g. with only 68.85%accuracy on COMPAS), adding an extra layer of protection on user privacy. Experiments validate our theoretical analyses and show our algorithm can effectively measure and mitigate bias. Our results imply a set of practical guidelines for practitioners on how to use proxies properly. Code is available at github.com/UCSC-REAL/fair-eval.

Jianing Qiu, Lin Li, Jiankai Sun et al.

Large AI models, or foundation models, are models recently emerging with massive scales both parameter-wise and data-wise, the magnitudes of which can reach beyond billions. Once pretrained, large AI models demonstrate impressive performance in various downstream tasks. A prime example is ChatGPT, whose capability has compelled people's imagination about the far-reaching influence that large AI models can have and their potential to transform different domains of our lives. In health informatics, the advent of large AI models has brought new paradigms for the design of methodologies. The scale of multi-modal data in the biomedical and health domain has been ever-expanding especially since the community embraced the era of deep learning, which provides the ground to develop, validate, and advance large AI models for breakthroughs in health-related areas. This article presents a comprehensive review of large AI models, from background to their applications. We identify seven key sectors in which large AI models are applicable and might have substantial influence, including 1) bioinformatics; 2) medical diagnosis; 3) medical imaging; 4) medical informatics; 5) medical education; 6) public health; and 7) medical robotics. We examine their challenges, followed by a critical discussion about potential future directions and pitfalls of large AI models in transforming the field of health informatics.

Jindong Hong, Tianjie Chen, Lingjie Luo et al.

A recent advancement in Multimodal Large Language Models (MLLMs) research is the emergence of "reasoning MLLMs" that offer explicit control over their internal thinking processes (normally referred as the "thinking mode") alongside the standard "non-thinking mode". This capability allows these models to engage in a step-by-step process of internal deliberation before generating a final response. With the rapid transition to and adoption of these "dual-state" MLLMs, this work rigorously evaluated how the enhanced reasoning processes of these MLLMs impact model performance and reliability in clinical tasks. This paper evaluates the active "thinking mode" capabilities of two leading MLLMs, Seed1.5-VL and Gemini-2.5-Flash, for medical applications. We assessed their performance on four visual medical tasks using VQA-RAD and ROCOv2 datasets. Our findings reveal that the improvement from activating the thinking mode remains marginal compared to the standard non-thinking mode for the majority of the tasks. Their performance on complex medical tasks such as open-ended VQA and medical image interpretation remains suboptimal, highlighting the need for domain-specific medical data and more advanced methods for medical knowledge integration.

Jianing Qiu, Lin Li, Jiankai Sun et al.

Large language models (LLMs)-powered AI agents exhibit a high level of autonomy in addressing medical and healthcare challenges. With the ability to access various tools, they can operate within an open-ended action space. However, with the increase in autonomy and ability, unforeseen risks also arise. In this work, we investigated one particular risk, i.e., cyber attack vulnerability of medical AI agents, as agents have access to the Internet through web browsing tools. We revealed that through adversarial prompts embedded on webpages, cyberattackers can: i) inject false information into the agent's response; ii) they can force the agent to manipulate recommendation (e.g., healthcare products and services); iii) the attacker can also steal historical conversations between the user and agent, resulting in the leak of sensitive/private medical information; iv) furthermore, the targeted agent can also cause a computer system hijack by returning a malicious URL in its response. Different backbone LLMs were examined, and we found such cyber attacks can succeed in agents powered by most mainstream LLMs, with the reasoning models such as DeepSeek-R1 being the most vulnerable.

Kungang Li, Xiangyi Chen, Ling Leng et al.

In the realm of online advertising, accurately predicting the conversion rate (CVR) is crucial for enhancing advertising efficiency and user satisfaction. This paper addresses the challenge of CVR prediction while adhering to user privacy preferences and advertiser requirements. Traditional methods face obstacles such as the reluctance of advertisers to share sensitive conversion data and the limitations of model training in secure environments like data clean rooms. We propose a novel model training framework that enables collaborative model training without sharing sample-level gradients with the advertising platform. Our approach introduces several innovative components: (1) utilizing batch-level aggregated gradients instead of sample-level gradients to minimize privacy risks; (2) applying adapter-based parameter-efficient fine-tuning and gradient compression to reduce communication costs; and (3) employing de-biasing techniques to train the model under label differential privacy, thereby maintaining accuracy despite privacy-enhanced label perturbations. Our experimental results, conducted on industrial datasets, demonstrate that our method achieves competitive ROCAUC performance while significantly decreasing communication overhead and complying with both advertiser privacy requirements and user privacy choices. This framework establishes a new standard for privacy-preserving, high-performance CVR prediction in the digital advertising landscape.

Xiaohui Chen, Jiankai Sun, Taiqing Wang et al.

Data subsampling is widely used to speed up the training of large-scale recommendation systems. Most subsampling methods are model-based and often require a pre-trained pilot model to measure data importance via e.g. sample hardness. However, when the pilot model is misspecified, model-based subsampling methods deteriorate. Since model misspecification is persistent in real recommendation systems, we instead propose model-agnostic data subsampling methods by only exploring input data structure represented by graphs. Specifically, we study the topology of the user-item graph to estimate the importance of each user-item interaction (an edge in the user-item graph) via graph conductance, followed by a propagation step on the network to smooth out the estimated importance value. Since our proposed method is model-agnostic, we can marry the merits of both model-agnostic and model-based subsampling methods. Empirically, we show that combing the two consistently improves over any single method on the used datasets. Experimental results on KuaiRec and MIND datasets demonstrate that our proposed methods achieve superior results compared to baseline approaches.

Weihao Gao, Xiangjun Fan, Chong Wang et al.

One of the core problems in large-scale recommendations is to retrieve top relevant candidates accurately and efficiently, preferably in sub-linear time. Previous approaches are mostly based on a two-step procedure: first learn an inner-product model, and then use some approximate nearest neighbor (ANN) search algorithm to find top candidates. In this paper, we present Deep Retrieval (DR), to learn a retrievable structure directly with user-item interaction data (e.g. clicks) without resorting to the Euclidean space assumption in ANN algorithms. DR's structure encodes all candidate items into a discrete latent space. Those latent codes for the candidates are model parameters and learnt together with other neural network parameters to maximize the same objective function. With the model learnt, a beam search over the structure is performed to retrieve the top candidates for reranking. Empirically, we first demonstrate that DR, with sub-linear computational complexity, can achieve almost the same accuracy as the brute-force baseline on two public datasets. Moreover, we show that, in a live production recommendation system, a deployed DR approach significantly outperforms a well-tuned ANN baseline in terms of engagement metrics. To the best of our knowledge, DR is among the first non-ANN algorithms successfully deployed at the scale of hundreds of millions of items for industrial recommendation systems.

Jiankai Sun, Jie Zhao, Huan Sun et al.

Routing newly posted questions (a.k.a cold questions) to potential answerers with the suitable expertise in Community Question Answering sites (CQAs) is an important and challenging task. The existing methods either focus only on embedding the graph structural information and are less effective for newly posted questions, or adopt manually engineered feature vectors that are not as representative as the graph embedding methods. Therefore, we propose to address the challenge of leveraging heterogeneous graph and textual information for cold question routing by designing an end-to-end framework that jointly learns CQA node embeddings and finds best answerers for cold questions. We conducted extensive experiments to confirm the usefulness of incorporating the textual information from question tags and demonstrate that an end-2-end framework can achieve promising performances on routing newly posted questions asked by both existing users and newly registered users.

Jiankai Sun, Srinivasan Parthasarathy

Recently, one has seen a surge of interest in developing such methods including ones for learning such representations for (undirected) graphs (while preserving important properties). However, most of the work to date on embedding graphs has targeted undirected networks and very little has focused on the thorny issue of embedding directed networks. In this paper, we instead propose to solve the directed graph embedding problem via a two-stage approach: in the first stage, the graph is symmetrized in one of several possible ways, and in the second stage, the so-obtained symmetrized graph is embedded using any state-of-the-art (undirected) graph embedding algorithm. Note that it is not the objective of this paper to propose a new (undirected) graph embedding algorithm or discuss the strengths and weaknesses of existing ones; all we are saying is that whichever be the suitable graph embedding algorithm, it will fit in the above proposed symmetrization framework.

Jiankai Sun, Bortik Bandyopadhyay, Armin Bashizade et al.

Directed graphs have been widely used in Community Question Answering services (CQAs) to model asymmetric relationships among different types of nodes in CQA graphs, e.g., question, answer, user. Asymmetric transitivity is an essential property of directed graphs, since it can play an important role in downstream graph inference and analysis. Question difficulty and user expertise follow the characteristic of asymmetric transitivity. Maintaining such properties, while reducing the graph to a lower dimensional vector embedding space, has been the focus of much recent research. In this paper, we tackle the challenge of directed graph embedding with asymmetric transitivity preservation and then leverage the proposed embedding method to solve a fundamental task in CQAs: how to appropriately route and assign newly posted questions to users with the suitable expertise and interest in CQAs. The technique incorporates graph hierarchy and reachability information naturally by relying on a non-linear transformation that operates on the core reachability and implicit hierarchy within such graphs. Subsequently, the methodology levers a factorization-based approach to generate two embedding vectors for each node within the graph, to capture the asymmetric transitivity. Extensive experiments show that our framework consistently and significantly outperforms the state-of-the-art baselines on two diverse real-world tasks: link prediction, and question difficulty estimation and expert finding in online forums like Stack Exchange. Particularly, our framework can support inductive embedding learning for newly posted questions (unseen nodes during training), and therefore can properly route and assign these kinds of questions to experts in CQAs.

Jiankai Sun, Sobhan Moosavi, Rajiv Ramnath et al.

In this paper, we present a framework for Question Difficulty and Expertise Estimation (QDEE) in Community Question Answering sites (CQAs) such as Yahoo! Answers and Stack Overflow, which tackles a fundamental challenge in crowdsourcing: how to appropriately route and assign questions to users with the suitable expertise. This problem domain has been the subject of much research and includes both language-agnostic as well as language conscious solutions. We bring to bear a key language-agnostic insight: that users gain expertise and therefore tend to ask as well as answer more difficult questions over time. We use this insight within the popular competition (directed) graph model to estimate question difficulty and user expertise by identifying key hierarchical structure within said model. An important and novel contribution here is the application of "social agony" to this problem domain. Difficulty levels of newly posted questions (the cold-start problem) are estimated by using our QDEE framework and additional textual features. We also propose a model to route newly posted questions to appropriate users based on the difficulty level of the question and the expertise of the user. Extensive experiments on real world CQAs such as Yahoo! Answers and Stack Overflow data demonstrate the improved efficacy of our approach over contemporary state-of-the-art models. The QDEE framework also allows us to characterize user expertise in novel ways by identifying interesting patterns and roles played by different users in such CQAs.