Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks78 from public data
- Federated Graph Neural Networks: Overview, Techniques, and Challenges205
A 2-D taxonomy of the FedGNN literature is proposed: the main taxonomy provides a clear perspective on the integration of GNNs and FL by analyzing how GNNs enhance FL training as well as how FL assists GNN training and the auxiliary taxonomy provides a view on how FedGNNs deal with heterogeneity across FL clients.
- 171
- Approximation Algorithms for Product Framing and Pricing84
One of the first models of “product framing” and pricing and pricing is proposed, based on approximation algorithms for product framing and pricing.
- Slideflow: deep learning for digital histopathology with real-time whole-slide visualization77
A flexible deep learning library for histopathology called Slideflow is developed, a package which supports a broad array of deep learning methods for digital pathology and includes a fast whole-slide interface for deploying trained models.
- Advances and Open Challenges in Federated Foundation Models75
A systematic multi-tiered taxonomy is proposed, categorizing existing FedFM approaches for model training, aggregation, trustworthiness, and incentivization, and offering a blueprint for future research directions, emphasizing the need for developing trustworthy solutions.
- FedMut: Generalized Federated Learning via Stochastic Mutation60
A novel FL approach is presented, which mutates the global model according to the gradient change to generate several intermediate models for the next round of training, which has a well-generalization compared with the global model trained by FedAvg.
- FedCross: Towards Accurate Federated Learning via Multi-Model Cross-Aggregation59
Experimental results on various well-known datasets show that, compared with state-of-the-art FL methods, Fed Cross can significantly improve FL accuracy within both IID and non-IID scenarios without causing additional communication overhead.
- 57
- FedSDG-FS: Efficient and Secure Feature Selection for Vertical Federated Learning50
The Federated Stochastic Dual-Gate based Feature Selection (FedSDG-FS) approach consists of a Gaussian stochastic dual-gate to efficiently approximate the probability of a feature being selected, with privacy protection through Partially Homomorphic Encryption without a trusted third-party.
- 40
- 39
- VulAdvisor: Natural Language Suggestion Generation for Software Vulnerability Repair31
VulAdvisor is introduced, an automated approach that generates natural language suggestions to guide developers or AVR tools in repairing vulnerabilities and it is shown that the generated suggestions enhance the patch generation capabilities of existing AVR tools.
- Ciron : a new benchmark dataset for Chinese irony detection25
Ciron is introduced, the first Chinese benchmark dataset available for irony detection for machine learning models, and is collected with no pre-conditions to ensure a much wider coverage.
- FlexFL: Heterogeneous Federated Learning via APoZ-Guided Flexible Pruning in Uncertain Scenarios24
A novel heterogeneous FL framework named FlexFL, based on the self-knowledge distillation, which can enhance the inference performance of the large models by learning the knowledge from the small models, can significantly improve the overall inference accuracy.
- 23
- 22
- 21
- 20
- FedPPO: Reinforcement Learning-Based Client Selection for Federated Learning With Heterogeneous Data19
- 19
- 19
- RF-Eye: Commodity RFID Can Know What You Write and Who You Are Wherever You Are17
RF-Eye is introduced, a system that enables contactless, position-independent handwriting recognition and user identification without prior training and proposes a novel feature, DCG, to capture changes in gesture direction and confirm its consistency across different positions.
- 17
- Learning Program Representations with a Tree-Structured Transformer14
The Tree-Transformer is proposed, a novel recursive tree-structured neural network to learn the vector representations for source codes that significantly outperforms the existing tree-based and graph-based program representation learning approaches in both the tree-level and node-level prediction tasks.
- FedCSS: Joint Client-and-Sample Selection for Hard Sample-Aware Noise-Robust Federated Learning14
The Federated Client and Sample Selection (FedCSS) approach is a bilevel optimization approach for FL client-and-sample selection to achieve hard sample-aware noise-robust learning in a privacy preserving manner.
- Residue-based Label Protection Mechanisms in Vertical Logistic Regression14
This paper first presents one label inference attack method to investigate the potential privacy leakages of the vertical logistic regression model, and proposes three protection mechanisms, e.g., additive noise mechanism, multiplicative noise mechanisms, and hybrid mechanism which leverages local differential privacy and homomorphic encryption techniques.
- 14
- Subnetwork-Lossless Robust Watermarking for Hostile Theft Attacks in Deep Transfer Learning Models13
A subnetwork-lossless robust DNN watermarking (SRDW) framework, which can exploit out-of-distribution guidance data augmentation to boost the robustness of watermarks and investigate how embedding watermarks in batch normalization layers affect the generalization performance of the deep transfer learning models, which reveals that reducing the embedding modifications in BN layers can further promote the robustity to resist hostile attacks.
- 13
- MultiSFL: Towards Accurate Split Federated Learning via Multi-Model Aggregation and Knowledge Replay12
A novel SFL approach named MultiSFL is proposed, which adopts an effective multi-model aggregation mechanism to alleviate gradient divergence caused by heterogeneous data and a novel knowledge replay strategy to deal with the catastrophic forgetting problem.
- An Empirical Study on Noisy Label Learning for Program Understanding12
An empirical study on the effectiveness of noisy label learning on deep learning for program understanding datasets, which evaluates various NLL approaches and deep learning models on three tasks: program classification, vulnerability detection, and code summarization.
- 12
- 11
- CaBaFL: Asynchronous Federated Learning via Hierarchical Cache and Feature Balance11
A novel asynchronous FL approach named CaBaFL, which includes a hierarchical cache-based aggregation mechanism and a feature balance-guided device selection strategy, which enables each intermediate model to be trained across devices with totally balanced data distributions before aggregation.
- 10
- Mechanistic Study on Transformation of Coamorphous Baicalein-Nicotinamide to Its Cocrystal Form10
This research discovered the mechanism of transformation from coamorphous BAI-NIC to cocrystal form, but also provided an unusual method for cocRYstal preparation from its coamorphously form.
- Fairness via Group Contribution Matching9
This study proposes an easy but effective group contribution matching (GCM) method to match the contribution of each subgroup in the training process, and shows that this method effectively improves fairness and outperforms other methods significantly.
- 9
- A Survey in Automatic Irony Processing: Linguistic, Cognitive, and Multi-X Perspectives9
This paper will provide a comprehensive overview of computational irony, insights from linguisic theory and cognitive science, as well as its interactions with downstream NLP tasks and newly proposed multi-X irony processing perspectives.
- 8
- 8
- 6
- 6
- 6
- 6
- A Method of Modern Chinese Irony Detection6
This research tries to find linguistic cues in grammatical and semantical level by the observation of large scale corpora to detect ironies, and concludes seven kinds of reversal in Chinese irony and summarize their formalized features.
- Historical Embedding-Guided Efficient Large-Scale Federated Graph Learning5
The Federated Adaptive Attention-based Sampling (FedAAS) approach achieves substantial cost savings by efficiently leveraging historical embedding estimators and focusing the limited communication resources on transmitting the most influential neighbor node embeddings across FL clients.
- 4
- 4
- LMOD \(\boldsymbol{+}\) : A Comprehensive Multimodal Dataset and Benchmark for Developing and Evaluating Multimodal Large Language Models in Ophthalmology3
A large-scale multimodal ophthalmology benchmark dataset comprising 32,633 instances with multi-granular annotations across 12 common ophthalmic conditions and 5 imaging modalities, and a systematic and unified data curation pipeline that repurposes existing or new datasets for MLLM development is presented.
- 3
- 3
- 3
- 2
- Federated Graph Learning with Adaptive Importance-based Sampling2
The Federated Adaptive Importance-based Sampling (FedAIS) approach achieves substantial computational cost saving by focusing the limited resources on training important nodes, while reducing communication overhead via adaptive historical embedding synchronization.
- 2
- Sustainable, Age-Friendly Cities: An Evaluative Case Study2
The research team developed a framework to systematically evaluate the age-friendliness of cities that included assessment using web-based Geographical Information System (GIS) tools and applied it to Palo Alto, Stanford as an illustrative case study.
- KoReA-SFL: Knowledge Replay-based Split Federated Learning Against Catastrophic Forgetting1
A novel SFL approach named KoReA-SFL is proposed, which adopts a multi-model aggregation mechanism to alleviate gradient divergence caused by heterogeneous data and a knowledge replay strategy to deal with catastrophic forgetting.
- 1
- 1
- The Power of Penalties: Negativity-Aware Incentives for High-Quality Crowdsourced Data Labeling–
This work establishes a paradigm shift in crowdsourcing system design, demonstrating that explicit negative modeling can enhance data quality, optimize costs, and foster participation at scale.
- –
- –
- –
- –
- –
- –
- Sewage Discharging in a Line: Global Optimization and Grand Cooperation–
This paper formalizes the sewage discharge problem (SDP) for different groups of players, and uses greedy strategy and dynamic programming to design the optimal algorithms to solve the SDP in polynomial time, and shows that the cooperative game defined on sewage discharged problem has a non-empty core due to its special line-positioning structure.
- –
- –
- Multiplicity of nontrivial solutions for Kirchhoff type equations with zero mass and a critical term–
- –
- The research on indirect incident detection in expressway under high traffic density–
The indirect incident detection method is designed and presented based on the actual data of the segment from the Lingdian interchange to the Tangwang interchange hub of the Jinan-Guangzhou expressway with an excellent performance on incident identification.
- –
- –
- A Study of Valence & Argument Integration in Chinese Verb-Resultative Complement–
This article has analyzed and summarized the existing research outcome of VC, and then gives further analyses thereby on argument integration process and multi-valence phenomena.
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.