STM Journals Peer-reviewed research from Consortium e-Learning Network

3 publications

  • Published Subscription Review Article

    Shell-Based Agents as Execution Systems: A Survey of Command Generation, Verification, Security, State, and Recovery

    By Jyoti Dabass, Bhupender Singh Dabass

    Abstract: Shell environments are increasingly becoming interfaces through which AI agents interact with software and operating systems. Unlike conventional command generation systems, shell-based agents can interpret task objectives, plan multi-step actions, invoke terminal tools, observe system state, and adapt subsequent actions. These capabilities support software development, system administration, experimentation, and automated operations, but also introduce reliability and security challenges that cannot be assessed through command generation alone. This survey reviews 57 …

    Published in Journal of Advances in Shell Programming · Vol. 13, Issue 2, 2026 · pp. 30–54 Read article

  • Published Subscription Original Research

    TRACE: Tracking the Loss of Memory Provenance in LLM Agents

    By Jyoti Dabass, Bhupender Singh Dabass

    Abstract: AI agents that carry memory across sessions gain better personalization and decision-making, but this persistence opens a serious security gap. When an agent repeatedly condenses earlier interactions into compact “lessons,” the trail linking each lesson back to the interaction that produced it gradually fades. We term this effect Reflective Attribution Collapse (RAC): the progressive loss of provenance and forensic traceability that results from repeated memory reflection. Under RAC, a malicious …

    Published in Journal of Advancements in Robotics · Vol. 13, Issue 2, 2026 · pp. 33–49 Read article

  • Published Subscription Review Article

    Stacked Generalization-Based Deep Learning Approach for Pneumonia Detection

    By Jyoti Dabass, Bhupender Singh Dabass

    Abstract: The proposed work focuses on a stacked generalization-based approach for diagnosing pneumonia from chest X-ray images. It utilizes regularization, early stopping, and data augmentation to deal with overfitting. It uses safe level SMOTE to deal with class imbalance and attention-based feature fusion to adaptively weigh features based on their importance. It uses two publicly available datasets (RSNA and Kermany) with ground truth provided by expert radiologists. The proposed work used …

    Published in Journal of Image Processing & Pattern Recognition Progress · Vol. 12, Issue 3, 2025 · pp. 20–31 Read article

Support