arXiv subjects

Dev Patel

Publications and source records attributed to Dev Patel.

6 recordsLinked to original sources

arXiv2026

"Stuck in a Spiral": Shame and Guilt as Social Regulators of AI Use in Computing Education

While prior work has examined patterns of adoption and social norms around AI use, less is known about how emotional factors, such as shame and guilt, shape students use of AI tools. We present an interview study with 19 computing students through a functionalist perspective of shame and guilt, which interprets emotions as social signals that regulate behavior. Our findings show that these emotions regulate when and how students make their use visible, as they engage in hiding behaviors and selective disclosure. Students described shaming themselves, their peers, and even faculty for using AI. Shame and guilt often coexist with continued AI use, creating cycles of reduced agency and moral tension rather than promoting behavior change. Students described feeling tensions between their AI use and their identities as competent, hardworking, or ethical computing students. Students also used language and metaphors of addiction to describe their experiences. These results highlight the need to consider the socio-emotional aspects of AI use, which may be influenced by how AI policies are implemented and enforced. We discuss classroom practices that can foster healthy, open discussion and support responsible AI use.

cs.CY↗

arXiv2025

Alignment-Constrained Dynamic Pruning for LLMs: Identifying and Preserving Alignment-Critical Circuits

Large Language Models require substantial computational resources for inference, posing deployment challenges. While dynamic pruning offers superior efficiency over static methods through adaptive circuit selection, it exacerbates alignment degradation by retaining only input-dependent safety-critical circuit preservation across diverse inputs. As a result, addressing these heightened alignment vulnerabilities remains critical. We introduce Alignment-Aware Probe Pruning (AAPP), a dynamic structured pruning method that adaptively preserves alignment-relevant circuits during inference, building upon Probe Pruning. Experiments on LLaMA 2-7B, Qwen2.5-14B-Instruct, and Gemma-3-12B-IT show AAPP improves refusal rates by 50\% at matched compute, enabling efficient yet safety-preserving LLM deployment.

cs.LG↗

arXiv2025

Hydra: A Modular Architecture for Efficient Long-Context Reasoning

The quadratic complexity of transformers fundamentally limits reasoning system deployment in resource-constrained and long-context settings. We introduce Hydra, a modular architecture based upon a state-space backbone which adaptively routes between complementary efficiency mechanisms: sparse global attention, mixture-of-experts, and dual memories comprising a reasoning workspace and product key memory. We evaluate a 29M parameter model measuring logical chaining accuracy and throughput on synthetic sequences, plus throughput on WikiText. Ablation studies use component-specific synthetic datasets to isolate individual mechanisms. Hydra achieves $3.01\times$ and $3.0\times$ throughput gains at 8K tokens for synthetic and WikiText datasets, respectively, and $10\times$ accuracy improvements on multi-step logical composition compared to equal-sized transformers. Ablations confirm each component's contribution: sparse attention captures long-range dependencies, experts specialize to input domains, and product key memory enables selective retrieval.

cs.LG↗

arXiv2024

Graph-Based Proofs of Indistinguishability of Linear Compartmental Models

Given experimental data, one of the main objectives of biological modeling is to construct a model which best represents the real world phenomena. In some cases, there could be multiple distinct models exhibiting the exact same dynamics, meaning from the modeling perspective it would be impossible to distinguish which model is ``correct.'' This is the study of indistinguishability of models, and in our case we focus on linear compartmental models which are often used to model pharmacokinetics, cell biology, ecology, and related fields. Specifically, we focus on a family of linear compartmental models called skeletal path models which have an underlying directed path, and have recently been shown to have the first recorded sufficient conditions for indistinguishability based on underlying graph structure. In this recent work, certain families of skeletal path models were proven to be indistinguishable, however the proofs relied heavily on linear algebra. In this work, we reprove several of these indistinguishability results instead using a graph theoretic framework.

math.CO↗

arXiv2020

On Aadhaar Identity Management System

A unique identification for citizens can lead to effective governance to manage and provide citizen-centric services. While ensuring this service, privacy of the citizens needs to be preserved. Aadhaar, the identification system by UIDAI has faced some critics regarding its privacy preserving feature. This paper discusses those concerns in Aadhaar system and proposed a new model for the Aadhaar system. The proposed solution is aimed to address the issue of collusion of third party service providers and profiling of Aadhaar users. The proposed solution uses a distributed model capturing the Aadhaar system, in which data of users is decentralized and stored in zonal office's databases as well as the CIDR. The proposed solution provides the functioning of the authentication process of the Aadhaar system more effective, as it reduces the number of requests being handled directly by the CIDR and also tackles the concern of correlation of data.

cs.CR↗

arXiv2018

Servo Actuating System Control Using Optimal Fuzzy Approach Based on Particle Swarm Optimization

This paper presents a new optimal fuzzy approach based on particle swarm optimization evolutionary algorithm for controlling the servo actuating system. It is clear that attaining the maximum stability margin is the prominent goal in control design of servo actuating systems. To reach the control goal, two main steps of design are required, an appropriate identification method and a controller development. Hence, the nonlinear system is first identified by the fuzzy algorithm. Then, the controller parameters and the algorithms weighting functions are tuned through the Particle Swarm Optimization algorithm. The objective function of optimal control strategy is such that the minimum error between the actual and the identified data is attained. The effectiveness of the proposed approach comparing to the conventional fuzzy control with regular parameter tuning is illustrated and analyzed in the simulations.

eess.SY↗