Large Language Models (LLMs) have transformed natural language processing, but face significant challenges in widespread deployment due to their high runtime cost. In this paper, we introduce SeedLM, a novel post-training compression method that uses seeds of a pseudo-random generator to encode and compress model weights. Specifically, for each block of weights, we find a seed that is fed into a Linear Feedback Shift Register (LFSR) during inference to efficiently generate a random matrix. This matrix is then linearly combined with compressed coefficients to reconstruct the weight block. SeedLM reduces memory access and leverages idle compute cycles during inference, effectively speeding up memory-bound tasks by trading compute for fewer memory accesses. Unlike state-of-the-art methods that rely on calibration data, our approach is data-free and generalizes well across diverse tasks. Our experiments with Llama3 70B, which is particularly challenging, show zero-shot accuracy retention at 4- and 3-bit compression to be on par with or better than state-of-the-art methods, while maintaining performance comparable to FP16 baselines. Additionally, FPGA-based tests demonstrate that 4-bit SeedLM, as model size increases, approaches a 4x speed-up over an FP16 Llama 2/3 baseline.

Abstract Virtual Production is a new process in the audiovisual industry, taking advantage of emerging technologies and attracting a significant degree of interest in academic and industrial research. This article documents a research process focusing on co-design of innovative solutions for Virtual Production relying on immersive technologies. Two remote collaborative workshops were organized involving audiovisual professionals covering different roles in different phases of the making process of audiovisual production. A range of innovative design concepts was generated as part of the research following group-based iterative discussion and evaluation. The study has contributed a set of innovative design solutions in relation to applications of immersive technologies in the audiovisual industry. The authors argue that the methods adopted have the potential to serve as a blueprint for design and implementation of future remote collaborative co-design processes in relation to audiovisual studies and, more generally, across disciplinary boundaries.

In common with many industries, the audiovisual sector is likely to be transformed by real-time graphic engines in combination with immersive technologies such as Virtual Reality and Augmented Reality. This technological mix enables what is presently known as Virtual Production, introducing a new process to the audiovisual professional, capable of fostering creativity, collaboration, and decision making, while at the same time increasing efficiency. The potential for Virtual Production to transform workflows and creative processes within large-scale productions in the audiovisual sector is significant and includes the prospective introduction of new capabilities for remote co-creation of content. However, barriers to democratisation need to be overcome, particularly in relation to adoption of the technology by small and medium independent productions that generally have limited access to resources and technical knowledge. Following an extensive study of the current literature and involvement in the research of professionals working in the field through online interviews, this article documents a first step towards filling this gap by investigating Virtual Production adoption scenarios for small and medium independent productions. The primary aim is to design intuitive, engaging, and effective solutions to address the needs of Directors, Cinematographers, and Producers. Thanks to the valuable time and experience of industry professionals, this study has gathered user requirements for a Virtual Production design solution capable of facilitating location scouting and pre-production. A novel framework for remote exploration of target locations within immersive environments, co-created with relevant stakeholders, is presented to lay the foundations of future co-design work.

Abstract This paper presents a comparative study of two artificial intelligent systems, namely; Multilayer Perceptron (MLP) and support vector machine (SVM), to classify six fault conditions and the normal (nonfaulty) condition of a centrifugal pump. A hybrid training method for MLP is proposed for this work based on the combination of Back Propagation (BP) and Genetic Algorithm (GA). The two training algorithms are tested and compared separately as well. Features are extracted using Discrete Wavelet Transform (DWT), both approximations, details, and two mother wavelets were used to investigate their effectiveness on feature extraction. GA is also used to optimize the number of hidden layers and neurons of MLP. In this study, the feature extraction, GA‐based hidden layers, neurons selection, training algorithm, and classification performance, based on the strengths and weaknesses of each method, are discussed. From the results obtained, it is observed that the DWT with both MLP‐BP and SVM produces better classification rates and performances.

Persuasive Technology has the potential to influence user behavior for social benefit, e.g. to reduce environmental impact, but designers are lacking guidance choosing among design techniques for influencing interaction.

We have developed a simple yet motivating course-project for use in a one-semester undergraduate course in Artificial Intelligence. The project is well-suited to small liberal arts college environments that lack specialized hardware, software, and faculty, fulfilling an unmet need for many faculty at such institutions.We describe the project and how to deploy and use supporting software for faculty wishing to adopt this project.

This paper introduces a hybrid server/P2P streaming system called bittorrent-assisted streaming system (BASS) for large-scale video-on-demand (VoD) services. By distributing the load among P2P connections as well as maintaining active server connections, BASS can increase the system scalability while decreasing media playout wait times. To analyze the benefits of BASS, we examine torrent trace data collected in the first week of distribution for Fedora Core 3 and develop an empirical model of bittorrent client performance. Based on this, we run trace-based simulations to evaluate BASS and show that it is more scalable than current unicast solutions and can greatly decrease the average waiting time before playback

This paper generalizes the TCP Vegas congestion avoidance mechanism and uses accumulation , buffered packets of a flow inside network routers, as a congestion measure based on which a family of congestion control schemes can be derived. We call this model Accumulation-based Congestion Control (ACC), which fits into the nonlinear optimization framework proposed by Kelly. The ACC model serves as a reference for packet-switching network implementations. We show that TCP Vegas is one possible scheme under this model. It is well known that Vegas suffers from round trip propagation delay estimation error and reverse path congestion. We therefore design a new Monaco scheme that solves these problems by employing an out-of-band, receiver-based accumulation estimator, with the support of two FIFO priority queues from the (congested) routers. Comparisons between these two schemes demonstrate that Monaco does not suffer from the problems mentioned above and achieves better performance than Vegas. We use ns-2 simulations and Linux implementation experiments to show that the static and dynamic performance of Monaco matches the theoretic results. One key issue regarding the ACC model in general, i.e., the scalability of bottleneck buffer requirement, and a solution using a virtual queueing algorithm are discussed and evaluated.

This paper generalizes the TCP Vegas congestion avoidance mechanism and proposes a model to use accumulation, buffered packets of a flow inside network routers, as a congestion measure on which a family of congestion control schemes can be derived. We call this model accumulation-based congestion control (ACC). We use a bit-by-bit fluid model to define the accumulation concept and develop a general control algorithm, which includes a set of control policies. The new prove its proportional fairness and global stability. The ACC model serves as a reference for packet network implementations. We show that TCP Vegas is one possible scheme, which fits into the ACC model. It is well known that Vegas suffer from round trip propagation delay estimation error and reverse path queuing delay. We therefore design a new scheme called Monaco, which solves these problems by employing an out-of-band receiver-based accumulation estimator, with minimal support from network routers. Analysis and simulations comparisons between Vegas and Monaco demonstrate the effectiveness of the Monaco accumulation estimator. We use ns-2 simulations to show that the static and dynamic performance of Monaco matches the theoretic results. One key issue regarding the ACC model in general, i.e., the scalability of router buffer requirement, is discussed.

The complexity and dynamics of the Internet is driving the demand for scalable and effective network control. This paper proposes a collaborative on-line simulation architecture to provide pro-active and automated control functions for networks. The general model includes autonomous on-line simulators which continuously monitor/model the network conditions and execute a search in the parameter state space for better settings of the protocol parameters. The protocol parameters are then tuned by the on-line simulation system. We describe the building blocks of this architecture and investigate the implementation challenges in the areas of network modeling, on-line simulation and parameter search. We also discuss the applicability of this system and present the simulation and test results of a preliminary implementation.

A scheme for programmable logic array (PLA) decomposition that consists of one level of PLAs followed by a second level of simple two-input logic gates is presented. The propagation delay is therefore the sum of the delay through one level of PLA and one level of two-input gates. Since the delay through a two-input gate is significantly less than that through a PLA, the timing performance of the new scheme is generally superior to those of earlier PLA decomposition schemes. The sizes of the PLAs used depend on the choice of the two-input gates. An algorithm is presented that chooses the functionality of the gates such that the areas of the first-level PLAs are minimized, further improving performance. The new decomposition scheme was developed for the automatic programming of a programmable logic device (PLD) which had basically a three level architecture. The functional unit for such a PLD is described and the application of the algorithm to the programming of these functional units is discussed. Experimental results show that the new scheme significantly reduces the area over the single PLA implementation.>

Article Free Access Share on A trace-driven analysis of the UNIX 4.2 BSD file system Authors: John K. Ousterhout Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CA Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CAView Profile , Hervé Da Costa Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CA Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CAView Profile , David Harrison Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CA Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CAView Profile , John A. Kunze Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CA Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CAView Profile , Mike Kupfer Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CA Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CAView Profile , James G. Thompson Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CA Computer Science Division, Electrical Engineering and Computer Sciences, University of California, Berkeley, CAView Profile Authors Info & Claims SOSP '85: Proceedings of the tenth ACM symposium on Operating systems principlesDecember 1985 Pages 15–24https://doi.org/10.1145/323647.323631Published:01 December 1985Publication History 390citation1,640DownloadsMetricsTotal Citations390Total Downloads1,640Last 12 Months127Last 6 weeks18 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF