arXiv subjects

Hoeseok Yang

Publications and source records attributed to Hoeseok Yang.

11 recordsLinked to original sources

arXiv2026

Dynamic Rowhammer Threshold Management: Temperature-Aware Threshold for In-DRAM Defenses

In-DRAM Rowhammer defenses pin the mitigation threshold at manufacture time, yet the true Rowhammer Threshold (TRHD) varies with runtime temperature. We propose Dynamic Rowhammer Threshold Management, a defense-agnostic runtime layer that re-sources each defense's threshold from the observed temperature once per epoch via a linear-$T$ model with a VRD-motivated guardband $g$, projecting the result onto SALT-C, PRAC, and TRR through each defense's threshold parameter. A decoupled oracle that scales physical TRHD per-DIMM by $\delta \sim \mathrm{N}(1, \sigma)$ breaks model self-consistency. The layer drives PRAC's 72 staleness breaches at 85$^\circ$C to zero; at $\sigma{=}0.10$, sweeping $g$ collapses PRAC breaches from 38.4 ($g{=}1.0$) to 9.6 ($g{=}0.9$). SALT-C drops from 10 nominal-static breaches to 2 (Dynamic) to 0 (bootstrap), at $\leq$5.1\% latency. TRR is capacity-limited; the layer acts as a diagnostic.

cs.AR↗

arXiv2026

Sensitivity-Aware Thresholding and Token Routing for Activation Sparsification in Large Language Models

Efficient inference in Large Language Models (LLMs) requires deciding where computation can be reduced while preserving model quality. We study this problem through multilayer perceptron (MLP) activation sparsification and token-level conditional routing. We first propose Sensitivity-Aware Thresholding for Sparsity (SATS), a threshold calibration method to choose layerwise gate thresholds using a local MLP output sensitivity proxy rather than calibrating thresholds directly from activation percentiles. While SATS retains the existing mechanism of sparsifying MLP activations by thresholding gate activations, it replaces percentile-based calibration with a sensitivity-aware selection rule. We then introduce a lightweight token routing framework that dynamically selects between a base path and a modified path on a per-token basis, rather than applying the modified computation uniformly to all tokens. We evaluate both methods on multiple recent open-weight LLMs. Our results show that SATS improves over the threshold-based sparsification baseline at matched actual sparsity and that token routing yields a more favorable quality-throughput trade-off than static activation modification baselines. Overall, our results suggest that improved threshold calibration and token routing can improve the quality-throughput trade-off in LLMs.

cs.LG↗

arXiv2026

Ensuring Deterministic Timing in a Federated GNSS Correction Pipeline with Lingua Franca

Embedded systems that combine hardware interrupts, buffering, and distributed communication are often perceived as inherently asynchronous and difficult to analyze. However, such systems can exhibit a deterministic timing structure when modeled using explicit logical-time semantics. This paper presents a Global Navigation Satellite System (GNSS) correction-data pipeline implemented as a federated Lingua Franca (LF) application. The federated LF program decomposes the end-to-end pipeline into reactors with explicit time semantics, including a time-triggered GNSS receiver, a UART interrupt stream derived from baud rate and First-In First-Out (FIFO) buffer characteristics, a periodic forwarding task, and downstream processing with jitter monitoring. Federated execution and runtime logs validate the analytically derived deterministic timing structure-including interrupt cadence, ring-buffer evolution, packetization behavior, and physical--logical jitter-yielding a reproducible and predictable timing profile.

cs.OS↗

arXiv2026

Visual Accommodation: Rethinking Image Scale as a Learnable Variable for Object Detection

We propose Ciliary-DETR (previous name: Elastic-DETR), a framework for test-time resolution adjustment analogous to biological accommodation. While multi-scale data augmentation improves robustness to scale variation, modern detectors rely on fixed inference resolutions, potentially limiting flexibility and robustness. Similar to the ciliary muscle, we introduce a lightweight scale predictor that dynamically estimates test-time scale factors across a wide range of input scales. The core challenge is that the optimal input scale is inherently unobservable under standard training setups. To address this challenge, we introduce a parametric formulation of desired scaling behavior, leading to loss-driven objectives that guide scale optimization. Overall, our method enables flexible and efficient single-pass inference, bridging the gap between training-time robustness and test-time adaptation.

cs.CV↗

arXiv2025

SparseInfer: Training-free Prediction of Activation Sparsity for Fast LLM Inference

Leveraging sparsity is crucial for optimizing large language model inference. however, modern LLMs employing SiLU as their activation function exhibit minimal activation sparsity. Recent research has proposed replacing SiLU with ReLU to induce significant activation sparsity and showed no downstream task accuracy degradation through fine tuning. However, taking full advantage of it required training a predictor to estimate this sparsity. In this paper, we introduce SparseInfer, a simple, light weight, and training free predictor for activation sparsity of ReLU field LLMs, in which activation sparsity is predicted by comparing only the sign bits of inputs and weights. To compensate for possible prediction inaccuracy, an adaptive tuning of the predictor's conservativeness is enabled, which can also serve as a control knob for optimizing LLM inference. The proposed method achieves approximately faster inference speed over the state of the art, with negligible accuracy loss of within 1%p.

cs.PF↗

arXiv2024

Simplifying Two-Stage Detectors for On-Device Inference in Remote Sensing

Deep learning has been successfully applied to object detection from remotely sensed images. Images are typically processed on the ground rather than on-board due to the computation power of the ground system. Such offloaded processing causes delays in acquiring target mission information, which hinders its application to real-time use cases. For on-device object detection, researches have been conducted on designing efficient detectors or model compression to reduce inference latency. However, highly accurate two-stage detectors still need further exploitation for acceleration. In this paper, we propose a model simplification method for two-stage object detectors. Instead of constructing a general feature pyramid, we utilize only one feature extraction in the two-stage detector. To compensate for the accuracy drop, we apply a high pass filter to the RPN's score map. Our approach is applicable to any two-stage detector using a feature pyramid network. In the experiments with state-of-the-art two-stage detectors such as ReDet, Oriented-RCNN, and LSKNet, our method reduced computation costs upto 61.2% with the accuracy loss within 2.1% on the DOTAv1.5 dataset. Source code will be released.

cs.CV↗

arXiv2024

DyRA: Portable Dynamic Resolution Adjustment Network for Existing Detectors

Achieving constant accuracy in object detection is challenging due to the inherent variability of object sizes. One effective approach to this problem involves optimizing input resolution, referred to as a multi-resolution strategy. Previous approaches to resolution optimization have often been based on pre-defined resolutions with manual selection. However, there is a lack of study on run-time resolution optimization for existing architectures. This paper introduces DyRA, a dynamic resolution adjustment network providing an image-specific scale factor for existing detectors. This network is co-trained with detectors utilizing specially designed loss functions, namely ParetoScaleLoss and BalanceLoss. ParetoScaleLoss determines an adaptive scale factor for robustness, while BalanceLoss optimizes overall scale factors according to the localization performance of the detector. The loss function is devised to minimize the accuracy drop across contrasting objectives of different-sized objects for scaling. Our proposed network can improve accuracy across various models, including RetinaNet, Faster-RCNN, FCOS, DINO, and H-Deformable-DETR. The code is available at https://github.com/DaEunFullGrace/DyRA.git.

cs.CV↗

arXiv2024

To Balance or to Not? Battery Aging-Aware Active Cell Balancing for Electric Vehicles

Due to manufacturing variabilities and temperature gradients within an electric vehicle's battery pack, the capacities of cells in it decrease differently over time. This reduces the usable capacity of the battery - the charge levels of one or more cells might be at the minimum threshold while most of the other cells have residual charge. Active cell balancing (i.e., transferring charge among cells) can equalize their charge levels, thereby increasing the battery pack's usable capacity. But performing balancing means additional charge transfer, which can result in energy loss and cell aging, akin to memory aging in storage technologies due to writing. This paper studies when cell balancing should be optimally triggered to minimize aging while maintaining the necessary driving capability. In particular, we propose optimization strategies for cell balancing while minimizing their impact on aging. By borrowing terminology from the storage domain, we refer to this as "wear leveling-aware" active balancing.

eess.SY↗

arXiv2023

Machine Learning-Driven Burrowing with a Snake-Like Robot

Subterranean burrowing is inherently difficult for robots because of the high forces experienced as well as the high amount of uncertainty in this domain. Because of the difficulty in modeling forces in granular media, we propose the use of a novel machine-learning control strategy to obtain optimal techniques for vertical self-burrowing. In this paper, we realize a snake-like bio-inspired robot that is equipped with an IMU and two triple-axis magnetometers. Utilizing magnetic field strength as an analog for depth, a novel deep learning architecture was proposed based on sinusoidal and random data in order to obtain a more efficient strategy for vertical self-burrowing. This strategy was able to outperform many other standard burrowing techniques and was able to automatically reach targeted burrowing depths. We hope these results will serve as a proof of concept for how optimization can be used to unlock the secrets of navigating in the subterranean world more efficiently.

cs.RO↗

arXiv2016

A Formal Approach to Power Optimization in CPSs with Delay-Workload Dependence Awareness

The design of cyber-physical systems (CPSs) faces various new challenges that are unheard of in the design of classical real-time systems. Power optimization is one of the major design goals that is witnessing such new challenges. The presence of interaction between the cyber and physical components of a CPS leads to dependence between the time delay of a computational task and the amount of workload in the next iteration. We demonstrate that it is essential to take this delay-workload dependence into consideration in order to achieve low power consumption. In this paper, we identify this new challenge, and present the first formal and comprehensive model to enable rigorous investigations on this topic. We propose a simple power management policy, and show that this policy achieves a best possible notion of optimality. In fact, we show that the optimal power consumption is attained in a "steady-state" operation and a simple policy of finding and entering this steady state suffices, which can be quite surprising considering the added complexity of this problem. Finally, we validated the efficiency of our policy with experiments.

cs.OH↗

arXiv2014

EURETILE D7.3 - Dynamic DAL benchmark coding, measurements on MPI version of DPSNN-STDP (distributed plastic spiking neural net) and improvements to other DAL codes

The EURETILE project required the selection and coding of a set of dedicated benchmarks. The project is about the software and hardware architecture of future many-tile distributed fault-tolerant systems. We focus on dynamic workloads characterised by heavy numerical processing requirements. The ambition is to identify common techniques that could be applied to both the Embedded Systems and HPC domains. This document is the first public deliverable of Work Package 7: Challenging Tiled Applications.

cs.DC↗