arXiv subjects

Mustafa Altun

Publications and source records attributed to Mustafa Altun.

5 recordsLinked to original sources

arXiv2026

Laguna M.1/XS.2 Technical Report

We present Laguna M.1 and Laguna XS.2, two Mixture-of-Experts foundation models built for long-horizon, agentic coding: M.1 has $225.8$B total parameters ($23.4$B activated per token) and XS.2 has $33.4$B total ($3$B activated). Both models were trained from scratch end-to-end inside the same internal system that we refer to as our Model Factory: a tightly-integrated stack of versioned data, training, evaluation, and inference components that turn model development into an industrial process. We describe the principles and design choices of the Model Factory and also detail the end-to-end training process of our models, throughout pre-training data and architecture, post-training stages, evaluation, and quantization. On agentic software engineering and terminal benchmarks (SWE-bench Verified, SWE-bench Multilingual, SWE-Bench Pro, and Terminal-Bench 2.0) M.1 and XS.2 are competitive with state-of-the-art open models in their respective weight classes. Laguna XS.2 weights are released under Apache~2.0 at https://huggingface.co/collections/poolside/laguna-xs2.

cs.AI↗

arXiv2021

Efficient Hardware Realizations of Feedforward Artificial Neural Networks

This article presents design techniques proposed for efficient hardware implementation of feedforward artificial neural networks (ANNs) under parallel and time-multiplexed architectures. To reduce their design complexity, after the weights of ANN are determined in a training phase, we introduce a technique to find the minimum quantization value used to convert the floating-point weight values to integers. For each design architecture, we also propose an algorithm that tunes the integer weights to reduce the hardware complexity avoiding a loss in the hardware accuracy. Furthermore, the multiplications of constant weights by input variables are implemented under the shift-adds architecture using the fewest number of addition/subtraction operations found by prominent previously proposed algorithms. Finally, we introduce a computer-aided design (CAD) tool, called SIMURG, that can describe an ANN design in hardware automatically based on the ANN structure and the solutions of proposed design techniques and algorithms. Experimental results indicate that the tuning techniques can significantly reduce the ANN hardware complexity under a design architecture and the multiplierless design of ANN can lead to a significant reduction in area and energy consumption, increasing the latency slightly.

cs.AR↗

arXiv2020

Memristive Learning Cellular Automata: Theory and Applications

Memristors are novel non volatile devices that manage to combine storing and processing capabilities in the same physical place.Their nanoscale dimensions and low power consumption enable the further design of various nanoelectronic processing circuits and corresponding computing architectures, like neuromorhpic, in memory, unconventional, etc.One of the possible ways to exploit the memristor's advantages is by combining them with Cellular Automata (CA).CA constitute a well known non von Neumann computing architecture that is based on the local interconnection of simple identical cells forming N-dimensional grids.These local interconnections allow the emergence of global and complex phenomena.In this paper, we propose a hybridization of the CA original definition coupled with memristor based implementation, and, more specifically, we focus on Memristive Learning Cellular Automata (MLCA), which have the ability of learning using also simple identical interconnected cells and taking advantage of the memristor devices inherent variability.The proposed MLCA circuit level implementation is applied on optimal detection of edges in image processing through a series of SPICE simulations, proving its robustness and efficacy.

cs.ET↗

arXiv2019

From Stochastic to Bit Stream Computing: Accurate Implementation of Arithmetic Circuits and Applications in Neural Networks

In this study, we propose a novel computing paradigm "Bit Stream Computing" that is constructed on the logic used in stochastic computing, but does not necessarily employ randomly or Binomially distributed bit streams as stochastic computing does. Any type of streams can be used either stochastic or deterministic. The proposed paradigm benefits from the area advantage of stochastic logic and the accuracy advantage of conventional binary logic. We implement accurate arithmetic multiplier and adder circuits, classified as asynchronous or synchronous; we also consider their suitability of processing successive streams. The proposed circuits are simulated both in gate level and in transistor level with AMS 0.35um CMOS technology to show the circuits' potential for practical use. We thoroughly compare the proposed adders and multipliers with their predecessors in the literature, individually and in a neural network application. Comparisons made in terms of area and accuracy clearly favor the proposed designs. We believe that this study opens up new horizons for computing that enables us to implement much smaller yet accurate arithmetic circuits compared to the conventional binary and stochastic ones.

cs.ET↗

arXiv2018

High Speed and Low Power Sensing Schemes for STT-MRAM with IPMTJs

STT-MRAM with interfacial-anisotropy-type perpendicular MTJ (IPMTJ) is a powerful candidate for the low switching energy design of STT-MRAM. In the literature, the reading operation of STT-MRAM structured with IPMTJs have been not studied until this time, in our knowledge. We investigated the reading operation of STT-MRAM structured with IPMTJs. To enumerate the read operations of the NVSenseAmp have successfully been performed a 2.5X reduction in average low power and a 13X increase in average high speed compared with the previous works.

cs.ET↗