In this paper, we introduce Selective-distillation for Class and Architecture-agnostic unleaRning (SCAR), a novel approximate unlearning method. SCAR efficiently eliminates specific information while preserving the model's test accuracy without using a retain set, which is a key component in state-of-the-art approximate unlearning algorithms. Our approach utilizes a modified Mahalanobis distance to guide the unlearning of the feature vectors of the instances to be forgotten, aligning them to the nearest wrong class distribution. Moreover, we propose a distillation-trick mechanism that distills the knowledge of the original model into the unlearning model with out-of-distribution images for retaining the original model's test performance without using any retain set. Importantly, we propose a self-forget version of SCAR that unlearns without having access to the forget set. We experimentally verified the effectiveness of our method, on three public datasets, comparing it with state-of-the-art methods. Our method obtains performance higher than methods that operate without the retain set and comparable w.r.t the best methods that rely on the retain set.
相關內容
This paper introduces the retrieval-augmented large language model with Definite Finite Automaton (DFA-RAG), a novel framework designed to enhance the capabilities of conversational agents using large language models (LLMs). Traditional LLMs face challenges in generating regulated and compliant responses in special scenarios with predetermined response guidelines, like emotional support and customer service. Our framework addresses these challenges by embedding a Definite Finite Automaton (DFA), learned from training dialogues, within the LLM. This structured approach acts as a semantic router which enables the LLM to adhere to a deterministic response pathway. The routing is achieved by the retrieval-augmentation generation (RAG) strategy, which carefully selects dialogue examples aligned with the current conversational context. The advantages of DFA-RAG include an interpretable structure through human-readable DFA, context-aware retrieval for responses in conversations, and plug-and-play compatibility with existing LLMs. Extensive benchmarks validate DFA-RAG's effectiveness, indicating its potential as a valuable contribution to the conversational agent.
We present NewsBench, a novel evaluation framework to systematically assess the capabilities of Large Language Models (LLMs) for editorial capabilities in Chinese journalism. Our constructed benchmark dataset is focused on four facets of writing proficiency and six facets of safety adherence, and it comprises manually and carefully designed 1,267 test samples in the types of multiple choice questions and short answer questions for five editorial tasks in 24 news domains. To measure performances, we propose different GPT-4 based automatic evaluation protocols to assess LLM generations for short answer questions in terms of writing proficiency and safety adherence, and both are validated by the high correlations with human evaluations. Based on the systematic evaluation framework, we conduct a comprehensive analysis of ten popular LLMs which can handle Chinese. The experimental results highlight GPT-4 and ERNIE Bot as top performers, yet reveal a relative deficiency in journalistic safety adherence in creative writing tasks. Our findings also underscore the need for enhanced ethical guidance in machine-generated journalistic content, marking a step forward in aligning LLMs with journalistic standards and safety considerations.
Recently, simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs) have emerged as a novel technology that provides 360 coverage and new degrees-of-freedom (DoFs). They are also capable of manipulating signal propagation and simultaneous wireless information and power transfer (SWIPT). This paper introduces a novel STAR-RIS-aided secure SWIPT system for downlink multiple input single output rate-splitting multiple access (RSMA) networks. The transmitter concurrently communicates with the information receivers (IRs) and sends energy to untrusted energy receivers (UERs). The UERs are also capable of wiretapping the IR streams. We assume that the channel state information (CSI) of the IRs is known at the information transmitter, but only imperfect CSI for the UERs is available at the energy transmitter. By exploiting RSMA, the base station splits the messages of the IRs into common and private parts. The former is encoded into a common stream that can be decoded by all IRs, while the private messages are individually decoded by their respective IRs. We find the precoders and STAR-RIS configuration that maximizes the achievable worst-case sum secrecy rate of the IRs under a total transmit power constraint, a sum energy constraint for the UERs, and subject to constraints on the transmission and reflection coefficients. The formulated problem is non-convex and has intricately coupled variables. To tackle this challenge, a suboptimal two-step iterative algorithm based on the sequential parametric convex approximation method is proposed. Simulations demonstrate that the RSMA-based algorithm implemented with a STAR-RIS enhances both the rate of confidential information transmission and the total spectral efficiency. Furthermore, our method surpasses the performance of both orthogonal multiple access (OMA) and non-OMA (NOMA).
In this paper, we propose a new set of midpoint-based high-order discretization schemes for computing straight and mixed nonlinear second derivative terms that appear in the compressible Navier-Stokes equations. Firstly, we detail a set of conventional fourth and sixth-order baseline schemes that utilize central midpoint derivatives for the calculation of second derivatives terms. To enhance the spectral properties of the baseline schemes, an optimization procedure is proposed that adjusts the order and truncation error of the midpoint derivative approximation while still constraining the same overall stencil width and scheme order. A new filter penalty term is introduced into the midpoint derivative calculation to help achieve high wavenumber accuracy and high-frequency damping in the mixed derivative discretization. Fourier analysis performed on the both straight and mixed second derivative terms show high spectral efficiency and minimal numerical viscosity with no odd-even decoupling effect. Numerical validation of the resulting optimized schemes is performed through various benchmark test cases assessing their theoretical order of accuracy and solution resolution. The results highlight that the present optimized schemes efficiently utilize the inherent viscosity of the governing equations to achieve improved simulation stability - a feature attributed to their superior spectral resolution in the high wavenumber range. The method is also tested and applied to non-uniform structured meshes in curvilinear coordinates, employing a supersonic impinging jet test case.
Decoupling the illumination in 3D scenes is crucial for novel view synthesis and relighting. In this paper, we propose a novel method for representing a scene illuminated by a point light using a set of relightable 3D Gaussian points. Inspired by the Blinn-Phong model, our approach decomposes the scene into ambient, diffuse, and specular components, enabling the synthesis of realistic lighting effects. To facilitate the decomposition of geometric information independent of lighting conditions, we introduce a novel bilevel optimization-based meta-learning framework. The fundamental idea is to view the rendering tasks under various lighting positions as a multi-task learning problem, which our meta-learning approach effectively addresses by generalizing the learned Gaussian geometries not only across different viewpoints but also across diverse light positions. Experimental results demonstrate the effectiveness of our approach in terms of training efficiency and rendering quality compared to existing methods for free-viewpoint relighting.
In this paper, we propose a novel data augmentation technique called GenMix, which combines generative and mixture approaches to leverage the strengths of both methods. While generative models excel at creating new data patterns, they face challenges such as mode collapse in GANs and difficulties in training diffusion models, especially with limited medical imaging data. On the other hand, mixture models enhance class boundary regions but tend to favor the major class in scenarios with class imbalance. To address these limitations, GenMix integrates both approaches to complement each other. GenMix operates in two stages: (1) training a generative model to produce synthetic images, and (2) performing mixup between synthetic and real data. This process improves the quality and diversity of synthetic data while simultaneously benefiting from the new pattern learning of generative models and the boundary enhancement of mixture models. We validate the effectiveness of our method on the task of classifying focal liver lesions (FLLs) in CT images. Our results demonstrate that GenMix enhances the performance of various generative models, including DCGAN, StyleGAN, Textual Inversion, and Diffusion Models. Notably, the proposed method with Textual Inversion outperforms other methods without fine-tuning diffusion model on the FLL dataset.
This paper consists of three parts. The first part provides a unified programming model for heterogeneous computing with CPU and accelerator (like GPU, FPGA, Google TPU, Atos QPU, and more) technologies. To some extent, this new programming model makes programming across CPUs and accelerators turn into usual programming tasks with common programming languages, and relieves complexity of programming across CPUs and accelerators. It can be achieved by extending file managements in common programming languages, such as C/C++, Fortran, Python, MPI, etc., to cover accelerators as I/O devices. In the second part, we show that all types of computer systems can be reduced to the simplest type of computer system, a single-core CPU computer system with I/O devices, by the unified programming model. Thereby, the unified programming model can truly build the programming of various computer systems on one API (i.e. file managements of common programming languages), and can make programming for various computer systems easier. In third part, we present a new approach to coupled applications computing (like multidisciplinary simulations) by the unified programming model. The unified programming model makes coupled applications computing more natural and easier since it only relies on its own power to couple multiple applications through MPI.
This article presents the affordances that Generative Artificial Intelligence can have in disinformation context, one of the major threats to our digitalized society. We present a research framework to generate customized agent-based social networks for disinformation simulations that would enable understanding and evaluation of the phenomena whilst discussing open challenges.
In this paper, we propose a novel Feature Decomposition and Reconstruction Learning (FDRL) method for effective facial expression recognition. We view the expression information as the combination of the shared information (expression similarities) across different expressions and the unique information (expression-specific variations) for each expression. More specifically, FDRL mainly consists of two crucial networks: a Feature Decomposition Network (FDN) and a Feature Reconstruction Network (FRN). In particular, FDN first decomposes the basic features extracted from a backbone network into a set of facial action-aware latent features to model expression similarities. Then, FRN captures the intra-feature and inter-feature relationships for latent features to characterize expression-specific variations, and reconstructs the expression feature. To this end, two modules including an intra-feature relation modeling module and an inter-feature relation modeling module are developed in FRN. Experimental results on both the in-the-lab databases (including CK+, MMI, and Oulu-CASIA) and the in-the-wild databases (including RAF-DB and SFEW) show that the proposed FDRL method consistently achieves higher recognition accuracy than several state-of-the-art methods. This clearly highlights the benefit of feature decomposition and reconstruction for classifying expressions.
In this paper, we proposed to apply meta learning approach for low-resource automatic speech recognition (ASR). We formulated ASR for different languages as different tasks, and meta-learned the initialization parameters from many pretraining languages to achieve fast adaptation on unseen target language, via recently proposed model-agnostic meta learning algorithm (MAML). We evaluated the proposed approach using six languages as pretraining tasks and four languages as target tasks. Preliminary results showed that the proposed method, MetaASR, significantly outperforms the state-of-the-art multitask pretraining approach on all target languages with different combinations of pretraining languages. In addition, since MAML's model-agnostic property, this paper also opens new research direction of applying meta learning to more speech-related applications.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191