In this paper, we introduce an algorithm for data quantization based on the principles of Kashin representation. This approach hinges on decomposing any given vector, matrix, or tensor into two factors. The first factor maintains a small infinity norm, while the second exhibits a similarly constrained norm when multiplied by an orthogonal matrix. Surprisingly, the entries of factors after decomposition are well-concentrated around several peaks, which allows us to efficiently replace them with corresponding centroids for quantization purposes. We study the theoretical properties of the proposed approach and rigorously evaluate our compression algorithm in the context of next-word prediction tasks and on a set of downstream tasks for text classification. Our findings demonstrate that Kashin Quantization achieves competitive or superior quality in model performance while ensuring data compression, marking a significant advancement in the field of data quantization.
相關內容
Synergizing Spatial Optimization with Large Language Models for Open-Domain Urban Itinerary Planning
In this paper, we introduce the novel task of Open-domain Urban Itinerary Planning (OUIP), a paradigm designed to generate personalized urban itineraries from user requests articulated in natural language. This approach is different from traditional itinerary planning, which often restricts the granularity of user inputs, thus hindering genuine personalization. To this end, we present ItiNera, an OUIP system that synergizes spatial optimization with large language models (LLMs) to provide services that customize urban itineraries based on users' needs. Upon receiving the user's itinerary request, the LLM first decomposes it into detailed components, identifying key requirements, including preferences and dislikes. Then, we use these specifics to select candidate POIs from a large-scale collection using embedding-based Preference-aware POI Retrieval. Finally, a preference score-based Cluster-aware Spatial Optimization module clusters, filters, and orders these POIs, followed by the LLM for detailed POI selection and organization to craft a personalized, spatially coherent itinerary. Moreover, we created an LLM-based pipeline to update and personalize a user-owned POI database. This ensures up-to-date POI information, supports itinerary planning, pre-trip research, POI collection, recommendations, and more. To the best of our knowledge, this study marks the first integration of LLMs to innovate itinerary planning, with potential extensions for various urban travel and exploration activities. Offline and online evaluations demonstrate the capacity of our system to deliver more responsive, personalized, and spatially coherent itineraries than current solutions. Our system, deployed on an online platform, has attracted thousands of users for their urban travel planning.
In this paper, we generalize the Jacobi eigenvalue algorithm to compute all eigenvalues and eigenvectors of a dual quaternion Hermitian matrix and show the convergence. We also propose a three-step Jacobi eigenvalue algorithm to compute the eigenvalues when a dual quaternion Hermitian matrix has two eigenvalues with identical standard parts but different dual parts and prove the convergence. Numerical experiments are presented to illustrate the efficiency and stability of the proposed Jacobi eigenvalue algorithm compaired to the power method and the Rayleigh quotient iteration method.
This paper explores the innovative application of Stable Video Diffusion (SVD), a diffusion model that revolutionizes the creation of dynamic video content from static images. As digital media and design industries accelerate, SVD emerges as a powerful generative tool that enhances productivity and introduces novel creative possibilities. The paper examines the technical underpinnings of diffusion models, their practical effectiveness, and potential future developments, particularly in the context of video generation. SVD operates on a probabilistic framework, employing a gradual denoising process to transform random noise into coherent video frames. It addresses the challenges of visual consistency, natural movement, and stylistic reflection in generated videos, showcasing high generalization capabilities. The integration of SVD in design tasks promises enhanced creativity, rapid prototyping, and significant time and cost efficiencies. It is particularly impactful in areas requiring frame-to-frame consistency, natural motion capture, and creative diversity, such as animation, visual effects, advertising, and educational content creation. The paper concludes that SVD is a catalyst for design innovation, offering a wide array of applications and a promising avenue for future research and development in the field of digital media and design.
In this paper, we propose low-complexity local detectors and log-likelihood ratio (LLR) refinement techniques for a coded cell-free massive multiple input multiple output (CF- mMIMO) systems, where an iterative detection and decoding (IDD) scheme is applied using parallel interference cancellation (PIC) and access point (AP) selection. In particular, we propose three LLR processing schemes based on the individual processing of the LLRs of each AP, LLR censoring, and a linear combination of LLRs by assuming statistical independence. We derive new closed-form expressions for the local soft minimum mean square error (MMSE)-PIC detector and receive matched filter (RMF). We also examine the system performance as the number of iterations increases. Simulations assess the performance of the proposed techniques against existing approaches.
The number of artificial intelligence algorithms for learning causal models from data is growing rapidly. Most ``causal discovery'' or ``causal structure learning'' algorithms are primarily validated through simulation studies. However, no widely accepted simulation standards exist and publications often report conflicting performance statistics -- even when only considering publications that simulate data from linear models. In response, several manuscripts have criticized a popular simulation design for validating algorithms in the linear case. We propose a new simulation design for generating linear models for directed acyclic graphs (DAGs): the DAG-adaptation of the Onion (DaO) method. DaO simulations are fundamentally different from existing simulations because they prioritize the distribution of correlation matrices rather than the distribution of linear effects. Specifically, the DaO method uniformly samples the space of all correlation matrices consistent with (i.e. Markov to) a DAG. We also discuss how to sample DAGs and present methods for generating DAGs with scale-free in-degree or out-degree. We compare the DaO method against two alternative simulation designs and provide implementations of the DaO method in Python and R: //github.com/bja43/DaO_simulation. We advocate for others to adopt DaO simulations as a fair universal benchmark.
Counterfactual explanations provide a popular method for analyzing the predictions of black-box systems, and they can offer the opportunity for computational recourse by suggesting actionable changes on how to change the input to obtain a different (i.e.\ more favorable) system output. However, recent work highlighted their vulnerability to different types of manipulations. This work studies the vulnerability of counterfactual explanations to data poisoning. We formally introduce and investigate data poisoning in the context of counterfactual explanations for increasing the cost of recourse on three different levels: locally for a single instance, or a sub-group of instances, or globally for all instances. In this context, we characterize and prove the correctness of several different data poisonings. We also empirically demonstrate that state-of-the-art counterfactual generation methods and toolboxes are vulnerable to such data poisoning.
In this paper, we investigate the problem of estimating the 4-DOF (three-dimensional position and orientation) robot-robot relative frame transformation using odometers and distance measurements between robots. Firstly, we apply a two-step estimation method based on maximum likelihood estimation. Specifically, a good initial value is obtained through unconstrained least squares and projection, followed by a more accurate estimate achieved through one-step Gauss-Newton iteration. Additionally, the optimal installation positions of Ultra-Wideband (UWB) are provided, and the minimum operating time under different quantities of UWB devices is determined. Simulation demonstrates that the two-step approach offers faster computation with guaranteed accuracy while effectively addressing the relative transformation estimation problem within limited space constraints. Furthermore, this method can be applied to real-time relative transformation estimation when a specific number of UWB devices are installed.
In this paper, we study the problem of uncertainty estimation and calibration for LLMs. We first formulate the uncertainty estimation problem for LLMs and then propose a supervised approach that takes advantage of the labeled datasets and estimates the uncertainty of the LLMs' responses. Based on the formulation, we illustrate the difference between the uncertainty estimation for LLMs and that for standard ML models and explain why the hidden neurons of the LLMs may contain uncertainty information. Our designed approach demonstrates the benefits of utilizing hidden activations to enhance uncertainty estimation across various tasks and shows robust transferability in out-of-distribution settings. We distinguish the uncertainty estimation task from the uncertainty calibration task and show that a better uncertainty estimation mode leads to a better calibration performance. Furthermore, our method is easy to implement and adaptable to different levels of model accessibility including black box, grey box, and white box.
This paper examines the complex nature of cyber attacks through an analysis of the LastPass breach. It argues for the integration of human-centric considerations into cybersecurity measures, focusing on mitigating factors such as goal-directed behavior, cognitive overload, human biases (e.g., optimism, anchoring), and risky behaviors. Findings from an analysis of this breach offers support to the perspective that addressing both the human and technical dimensions of cyber defense can significantly enhance the resilience of cyber systems against complex threats. This means maintaining a balanced approach while simultaneously simplifying user interactions, making users aware of biases, and discouraging risky practices are essential for preventing cyber incidents.
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.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191