Multimodal news contains a wealth of information and is easily affected by deepfake modeling attacks. To combat the latest image and text generation methods, we present a new Multimodal Fake News Detection dataset (MFND) containing 11 manipulated types, designed to detect and localize highly authentic fake news. Furthermore, we propose a Shallow-Deep Multitask Learning (SDML) model for fake news, which fully uses unimodal and mutual modal features to mine the intrinsic semantics of news. Under shallow inference, we propose the momentum distillation-based light punishment contrastive learning for fine-grained uniform spatial image and text semantic alignment, and an adaptive cross-modal fusion module to enhance mutual modal features. Under deep inference, we design a two-branch framework to augment the image and text unimodal features, respectively merging with mutual modalities features, for four predictions via dedicated detection and localization projections. Experiments on both mainstream and our proposed datasets demonstrate the superiority of the model. Codes and dataset are released at https://github.com/yunan-wang33/sdml.

Reinforcement learning (RL) agents are vulnerable to adversarial disturbances, which can deteriorate task performance or break down safety specifications. Existing methods either address safety requirements under the assumption of no adversary (e.g., safe RL) or only focus on robustness against performance adversaries (e.g., robust RL). Learning one policy that is both safe and robust under any adversaries remains a challenging open problem. The difficulty is how to tackle two intertwined aspects in the worst cases: feasibility and optimality. The optimality is only valid inside a feasible region (i.e., robust invariant set), while the identification of maximal feasible region must rely on how to learn the optimal policy. To address this issue, we propose a systematic framework to unify safe RL and robust RL, including the problem formulation, iteration scheme, convergence analysis and practical algorithm design. The unification is built upon constrained two-player zero-sum Markov games, in which the objective for protagonist is twofold. For states inside the maximal robust invariant set, the goal is to pursue rewards under the condition of guaranteed safety; for states outside the maximal robust invariant set, the goal is to reduce the extent of constraint violation. A dual policy iteration scheme is proposed, which simultaneously optimizes a task policy and a safety policy. We prove that the iteration scheme converges to the optimal task policy which maximizes the twofold objective in the worst cases, and the optimal safety policy which stays as far away from the safety boundary. The convergence of safety policy is established by exploiting the monotone contraction property of safety self-consistency operators, and that of task policy depends on the transformation of safety constraints into state-dependent action spaces. By adding two adversarial networks (one is for safety guarantee and the other is for task performance), we propose a practical deep RL algorithm for constrained zero-sum Markov games, called dually robust actor-critic (DRAC). The evaluations with safety-critical benchmarks demonstrate that DRAC achieves high performance and persistent safety under all scenarios (no adversary, safety adversary, performance adversary), outperforming all baselines by a large margin.

Hybrid additive-subtractive manufacturing (HASM) has achieved universal application in aerospace, medicine, military industry, and other fields. The additive manufacturing (AM) process has a significant influence on manufacturing quality and efficiency in HASM. However, few process planning methods exist utilizing the interplay between different processes in HASM. For the first time to the best of our knowledge, this paper proposes a slice extension (SE) method for HASM, the key idea of which is to print an extended slice in the AM process and guarantee the final geometric accuracy in the subtractive manufacturing (SM) process. Firstly, an optimization problem minimizing the isoperimetric quotient, the area, and the perimeter of the extended slice, is designed and convexified. Then, a unilateral rolling circle (URC) method is proposed to guarantee that the extended slice covers the original slice. Finally, an approach to dealing with multiply connected slices is proposed to eliminate underfill near holes. Simulation experiments indicate that the proposed SE method achieved an 81.5% lower underfill rate than the conventional method, i.e., the widely-used contour-parallel (CP) method in the industry, with limited extra materials on average. The toolpath smoothness is also improved. The performance of the proposed SE method verifies its significant application potential in HASM, computer numerical control milling, and other fields.

Electronic Health Record (EHR) contains patient-centric, electronically maintained information about individuals health information. Data Visualization that integrates with EHR is helpful for clinicians and patients to gain better insight into clinical care and management. This paper focuses on building EHR, and it contains two aspects: 1. 3D deformable human body model is applied as the health data visualization method, it focuses on human data visualization that is available for constructing customized digital human model. The model contains body surface, organs, muscle and skeleton, and is changeable together with users' real health data. 2. A design flow for automatically digitized the paper-based medical record to complete personal EHR through pictures that are obtained directly from mobile devices. The method works excellent on correcting errors happened during digitized, especially for the records with photographed, printed, and stored problems. The method achieve a 100% accuracy in digitization, tested using 200 medical reports with over 2000 pages.