Source author record

Ming-Syan Chen

Ming-Syan Chen appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Social and Information Networks Artificial Intelligence Computer Vision Cryptography and Security Data Structures and Algorithms Machine Learning physics.soc-ph

Catalog footprint

What is connected

12works

7topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Amortized-Precision Quantization for Early-Exit Vision Transformers

Vision Transformers (ViTs) achieve strong performance across vision tasks, yet their deployment with low-precision early exiting remains fragile. Existing quantization methods assume static full-depth execution, making them unstable when exit decisions are perturbed by quantization noise, which can amplify errors along dynamic inference paths. In this paper, we introduce Amortized-Precision Quantization (APQ), a utilization-aware formulation that accounts for layer-wise stochastic exposure to quantization noise and reveals depth-precision trade-offs. Building on APQ, we propose Mutual Adaptive Quantization with Early Exiting (MAQEE), a bi-level framework that jointly optimizes exit thresholds and bit-widths under explicit risk control to improve inference stability. MAQEE establishes a superior Pareto frontier in the accuracy-efficiency trade-off, reducing BOPs by up to 95% while maintaining accuracy and outperforming strong baselines by up to 20\% across classification, detection, and segmentation tasks.

preprint2026arXiv

Interference-Aware Multi-Task Unlearning

Machine unlearning aims to remove the contribution of designated training data from a trained model while preserving performance on the remaining data. Existing work mainly focuses on single-task settings, whereas modern models often operate in multi-task setups with shared backbones, where removing supervision for one task or instance can unintentionally affect others. We introduce multi-task unlearning with two settings: full-task unlearning, which removes a target instance from all tasks, and partial-task unlearning, which removes supervision only from selected tasks. We show that shared parameters couple the forget and retain sets, causing task-level interference on non-target tasks and instance-level interference on other instances. To address this issue, we propose an interference-aware framework that combines task-aware gradient projection, which constrains updates within task-specific subspaces, with instance-level gradient orthogonalization, which reduces conflicts between forget and retain signals. Experiments on two multi-task computer vision benchmarks across five tasks show that our method achieves effective unlearning while maintaining strong generalization, reducing UIS compared with the strongest baseline by 30.3% in full-task unlearning and 52.9% in partial-task unlearning.

preprint2026arXiv

LoopQ: Quantization for Recursive Transformers

Looped language models (LoopLMs) improve parameter efficiency by recursively reusing Transformer blocks, enabling deeper computation under a fixed model size. However, this reuse makes LoopLMs more fragile under post-training quantization (PTQ). We present the first systematic study of quantization in LoopLMs and identify three challenges: distribution shift across roles, state reuse across loop transitions, and recursive error accumulation. To address these challenges, we propose LoopQ, a loop-aware PTQ framework that preserves a shared quantized backbone while introducing lightweight adaptations. LoopQ combines activation scaling, selective transformation, cross-loop state alignment, and trajectory-aware optimization to reduce distributional mismatch within loops and error accumulation across loops. Experiments across seven benchmarks show that, under W4A4 quantization, LoopQ improves average downstream accuracy by 68.8% and reduces average perplexity by 87.7% compared with the strongest static PTQ baseline.

preprint2022arXiv

PGADA: Perturbation-Guided Adversarial Alignment for Few-shot Learning Under the Support-Query Shift

Few-shot learning methods aim to embed the data to a low-dimensional embedding space and then classify the unseen query data to the seen support set. While these works assume that the support set and the query set lie in the same embedding space, a distribution shift usually occurs between the support set and the query set, i.e., the Support-Query Shift, in the real world. Though optimal transportation has shown convincing results in aligning different distributions, we find that the small perturbations in the images would significantly misguide the optimal transportation and thus degrade the model performance. To relieve the misalignment, we first propose a novel adversarial data augmentation method, namely Perturbation-Guided Adversarial Alignment (PGADA), which generates the hard examples in a self-supervised manner. In addition, we introduce Regularized Optimal Transportation to derive a smooth optimal transportation plan. Extensive experiments on three benchmark datasets manifest that our framework significantly outperforms the eleven state-of-the-art methods on three datasets.

preprint2016arXiv

When Social Influence Meets Item Inference

Research issues and data mining techniques for product recommendation and viral marketing have been widely studied. Existing works on seed selection in social networks do not take into account the effect of product recommendations in e-commerce stores. In this paper, we investigate the seed selection problem for viral marketing that considers both effects of social influence and item inference (for product recommendation). We develop a new model, Social Item Graph (SIG), that captures both effects in form of hyperedges. Accordingly, we formulate a seed selection problem, called Social Item Maximization Problem (SIMP), and prove the hardness of SIMP. We design an efficient algorithm with performance guarantee, called Hyperedge-Aware Greedy (HAG), for SIMP and develop a new index structure, called SIG-index, to accelerate the computation of diffusion process in HAG. Moreover, to construct realistic SIG models for SIMP, we develop a statistical inference based framework to learn the weights of hyperedges from data. Finally, we perform a comprehensive evaluation on our proposals with various baselines. Experimental result validates our ideas and demonstrates the effectiveness and efficiency of the proposed model and algorithms over baselines.

preprint2015arXiv

An Effective Marketing Strategy for Revenue Maximization with a Quantity Constraint

Recently the influence maximization problem has received much attention for its applications on viral marketing and product promotions. However, such influence maximization problems have not taken into account the monetary effect on the purchasing decision of individuals. To fulfill this gap, in this paper, we aim for maximizing the revenue by considering the quantity constraint on the promoted commodity. For this problem, we not only identify a proper small group of individuals as seeds for promotion but also determine the pricing of the commodity. To tackle the revenue maximization problem, we first introduce a strategic searching algorithm, referred to as Algorithm PRUB, which is able to derive the optimal solutions. After that, we further modify PRUB to propose a heuristic, Algorithm PRUB+IF, for obtaining feasible solutions more effciently on larger instances. Experiments on real social networks with different valuation distributions demonstrate the effectiveness of PRUB and PRUB+IF.

preprint2015arXiv

Maximizing Friend-Making Likelihood for Social Activity Organization

The social presence theory in social psychology suggests that computer-mediated online interactions are inferior to face-to-face, in-person interactions. In this paper, we consider the scenarios of organizing in person friend-making social activities via online social networks (OSNs) and formulate a new research problem, namely, Hop-bounded Maximum Group Friending (HMGF), by modeling both existing friendships and the likelihood of new friend making. To find a set of attendees for socialization activities, HMGF is unique and challenging due to the interplay of the group size, the constraint on existing friendships and the objective function on the likelihood of friend making. We prove that HMGF is NP-Hard, and no approximation algorithm exists unless P = NP. We then propose an error-bounded approximation algorithm to efficiently obtain the solutions very close to the optimal solutions. We conduct a user study to validate our problem formulation and per- form extensive experiments on real datasets to demonstrate the efficiency and effectiveness of our proposed algorithm.

preprint2015arXiv

Scale-Adaptive Group Optimization for Social Activity Planning

Studies have shown that each person is more inclined to enjoy a group activity when 1) she is interested in the activity, and 2) many friends with the same interest join it as well. Nevertheless, even with the interest and social tightness information available in online social networks, nowadays many social group activities still need to be coordinated manually. In this paper, therefore, we first formulate a new problem, named Participant Selection for Group Activity (PSGA), to decide the group size and select proper participants so that the sum of personal interests and social tightness of the participants in the group is maximized, while the activity cost is also carefully examined. To solve the problem, we design a new randomized algorithm, named Budget-Aware Randomized Group Selection (BARGS), to optimally allocate the computation budgets for effective selection of the group size and participants, and we prove that BARGS can acquire the solution with a guaranteed performance bound. The proposed algorithm was implemented in Facebook, and experimental results demonstrate that social groups generated by the proposed algorithm significantly outperform the baseline solutions.

preprint2013arXiv

Structural Diversity for Resisting Community Identification in Published Social Networks

As an increasing number of social networking data is published and shared for commercial and research purposes, privacy issues about the individuals in social networks have become serious concerns. Vertex identification, which identifies a particular user from a network based on background knowledge such as vertex degree, is one of the most important problems that has been addressed. In reality, however, each individual in a social network is inclined to be associated with not only a vertex identity but also a community identity, which can represent the personal privacy information sensitive to the public, such as political party affiliation. This paper first addresses the new privacy issue, referred to as community identification, by showing that the community identity of a victim can still be inferred even though the social network is protected by existing anonymity schemes. For this problem, we then propose the concept of \textit{structural diversity} to provide the anonymity of the community identities. The $k$-Structural Diversity Anonymization ($k$-SDA) is to ensure sufficient vertices with the same vertex degree in at least $k$ communities in a social network. We propose an Integer Programming formulation to find optimal solutions to $k$-SDA and also devise scalable heuristics to solve large-scale instances of $k$-SDA from different perspectives. The performance studies on real data sets from various perspectives demonstrate the practical utility of the proposed privacy scheme and our anonymization approaches.

preprint2013arXiv

Willingness Optimization for Social Group Activity

Studies show that a person is willing to join a social group activity if the activity is interesting, and if some close friends also join the activity as companions. The literature has demonstrated that the interests of a person and the social tightness among friends can be effectively derived and mined from social networking websites. However, even with the above two kinds of information widely available, social group activities still need to be coordinated manually, and the process is tedious and time-consuming for users, especially for a large social group activity, due to complications of social connectivity and the diversity of possible interests among friends. To address the above important need, this paper proposes to automatically select and recommend potential attendees of a social group activity, which could be very useful for social networking websites as a value-added service. We first formulate a new problem, named Willingness mAximization for Social grOup (WASO). This paper points out that the solution obtained by a greedy algorithm is likely to be trapped in a local optimal solution. Thus, we design a new randomized algorithm to effectively and efficiently solve the problem. Given the available computational budgets, the proposed algorithm is able to optimally allocate the resources and find a solution with an approximation ratio. We implement the proposed algorithm in Facebook, and the user study demonstrates that social groups obtained by the proposed algorithm significantly outperform the solutions manually configured by users.

preprint2012arXiv

Implementation of Privacy-preserving SimRank over Distributed Information Network

Information network analysis has drawn a lot attention in recent years. Among all the aspects of network analysis, similarity measure of nodes has been shown useful in many applications, such as clustering, link prediction and community identification, to name a few. As linkage data in a large network is inherently sparse, it is noted that collecting more data can improve the quality of similarity measure. This gives different parties a motivation to cooperate. In this paper, we address the problem of link-based similarity measure of nodes in an information network distributed over different parties. Concerning the data privacy, we propose a privacy-preserving SimRank protocol based on fully-homomorphic encryption to provide cryptographic protection for the links.

preprint2011arXiv

On Social-Temporal Group Query with Acquaintance Constraint

Three essential criteria are important for activity planning, including: (1) finding a group of attendees familiar with the initiator, (2) ensuring each attendee in the group to have tight social relations with most of the members in the group, and (3) selecting an activity period available for all attendees. Therefore, this paper proposes Social-Temporal Group Query to find the activity time and attendees with the minimum total social distance to the initiator. Moreover, this query incorporates an acquaintance constraint to avoid finding a group with mutually unfamiliar attendees. Efficient processing of the social-temporal group query is very challenging. We show that the problem is NP-hard via a proof and formulate the problem with Integer Programming. We then propose two efficient algorithms, SGSelect and STGSelect, which include effective pruning techniques and employ the idea of pivot time slots to substantially reduce the running time, for finding the optimal solutions. Experimental results indicate that the proposed algorithms are much more efficient and scalable. In the comparison of solution quality, we show that STGSelect outperforms the algorithm that represents manual coordination by the initiator.

Institution

Affiliation not imported yet

This author record came from a source that does not expose affiliation metadata. Once the author claims the profile or we enrich the record from another provider, this section will link to the concrete institution.

Topic footprint