Anda Cheng
YOU?
Author Swipe
View article: DPF-CM: A Data Processing Framework with Privacy-Preserving Vector Databases for Chinese Medical LLMs Training and Deployment
DPF-CM: A Data Processing Framework with Privacy-Preserving Vector Databases for Chinese Medical LLMs Training and Deployment Open
Current open-source training pipelines for Chinese medical language models predominantly emphasize optimizing training methodologies to enhance the performance of large language models (LLMs), yet lack comprehensive exploration into traini…
View article: CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models
CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models Open
Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by incorporating external knowledge, but its openness introduces vulnerabilities that can be exploited by poisoning attacks. Existing poisoning methods for RAG syst…
View article: Information Leakage from Embedding in Large Language Models
Information Leakage from Embedding in Large Language Models Open
The widespread adoption of large language models (LLMs) has raised concerns regarding data privacy. This study aims to investigate the potential for privacy invasion through input reconstruction attacks, in which a malicious model provider…
View article: A Fast, Performant, Secure Distributed Training Framework For Large Language Model
A Fast, Performant, Secure Distributed Training Framework For Large Language Model Open
The distributed (federated) LLM is an important method for co-training the domain-specific LLM using siloed data. However, maliciously stealing model parameters and data from the server or client side has become an urgent problem to be sol…
View article: HPN: Personalized Federated Hyperparameter Optimization
HPN: Personalized Federated Hyperparameter Optimization Open
Numerous research studies in the field of federated learning (FL) have attempted to use personalization to address the heterogeneity among clients, one of FL's most crucial and challenging problems. However, existing works predominantly fo…
View article: PKD: General Distillation Framework for Object Detectors via Pearson Correlation Coefficient
PKD: General Distillation Framework for Object Detectors via Pearson Correlation Coefficient Open
Knowledge distillation(KD) is a widely-used technique to train compact models in object detection. However, there is still a lack of study on how to distill between heterogeneous detectors. In this paper, we empirically find that better FP…
View article: DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy
DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy Open
Training deep neural networks (DNNs) for meaningful differential privacy (DP) guarantees severely degrades model utility. In this paper, we demonstrate that the architecture of DNNs has a significant impact on model utility in the context …
View article: Differentially Private Federated Learning with Local Regularization and Sparsification
Differentially Private Federated Learning with Local Regularization and Sparsification Open
User-level differential privacy (DP) provides certifiable privacy guarantees to the information that is specific to any user's data in federated learning. Existing methods that ensure user-level DP come at the cost of severe accuracy decre…
View article: DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy
DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy Open
Training deep neural networks (DNNs) for meaningful differential privacy (DP) guarantees severely degrades model utility. In this paper, we demonstrate that the architecture of DNNs has a significant impact on model utility in the context …
View article: Location-aware Upsampling for Semantic Segmentation
Location-aware Upsampling for Semantic Segmentation Open
Many successful learning targets such as minimizing dice loss and cross-entropy loss have enabled unprecedented breakthroughs in segmentation tasks. Beyond these semantic metrics, this paper aims to introduce location supervision into sema…
View article: SpatialFlow: Bridging All Tasks for Panoptic Segmentation
SpatialFlow: Bridging All Tasks for Panoptic Segmentation Open
Object location is fundamental to panoptic segmentation as it is related to all things and stuff in the image scene. Knowing the locations of objects in the image provides clues for segmenting and helps the network better understand the sc…