Research

This research paper examines why classical empirical risk minimization (ERM) performs well on open set domain adaptation (OSDA), a setting with unknown target classes and shifted distributions. The authors introduce the fully informative causal invariance model (FICIM) and the partially informative causal invariance model (PICIM), and derive an OSDA theoretical bound showing ERM performs well under FICIM but poorly under PICIM. Experiments on multiple datasets validate these results, which the authors suggest may also inform training and fine-tuning of large language models (LLMs).

This paper presents NatIMG-FL, a federated learning framework for classifying X-ray images across healthcare institutions without sharing sensitive data. It uses natural images as auxiliary supervised data to align feature distributions between natural and X-ray images, and introduces a dual weights-based knowledge transfer method to handle heterogeneous model architectures.

FedWeight is a density-aware adaptive weight optimization method for federated learning that targets data heterogeneity across clients with non-IID data. Each client adjusts its loss function weights per class, using density estimated by kernel density estimation, and assigns larger weights to sparse classes. The authors report theoretical analysis and experiments showing more accurate models and faster convergence than integrated baseline approaches.

The paper proposes Feature-space Planes Searcher (FPS), an unsupervised domain adaptation framework that corrects decision-boundary misalignment while keeping a pre-trained feature encoder frozen. It argues this avoids the cost and unpredictable distortions of fine-tuning, and adds an Intra-Class Distance Metric (ICDM) for label-free hyperparameter selection. The authors report competitive benchmark results and applications including protein structure prediction, remote sensing classification and earthquake detection.

Researchers examined whether controller actions at the joint level of a robot can be inferred from encrypted network traffic. The source text provided is limited to the title, publication details and authors, so the method and findings are not described.

Researchers propose a triple-chain architecture for sensitive thumbnail-preserving encryption (TPE), arguing that existing TPE schemes encrypt pixels, blocks, or channels independently, which creates security vulnerabilities. The design links three encryption chains at the pixel, block, and channel levels, so any minor change to the original or encrypted image produces completely different encryption or decryption results. The authors report that the approach achieves sensitive TPE while keeping visual usability and privacy protection.

This paper introduces CLLME, a certificateless lattice-based matchmaking encryption scheme for bilateral access control in data sharing. Both senders and receivers specify matching access structures, so decryption requires mutual authorization, and a lightweight authenticity check filters out irrelevant ciphertexts before costly decryption. The authors prove indistinguishability against chosen-plaintext attacks and existential unforgeability against chosen-message attacks under lattice-based assumptions, and report favorable communication and computation efficiency.

The paper proposes Xsyn, a one-stage X-ray security image synthesis pipeline built on text-to-image generation, to avoid the labor-intensive foreground extraction of prior two-stage methods. It adds Cross-Attention Refinement (CAR), which uses the diffusion model's cross-attention map to refine bounding box annotations, and Background Occlusion Modeling (BOM), which models background occlusion in latent space. Experiments report a 1.2% mAP improvement over previous methods and better prohibited item detection across several X-ray datasets and detectors.