Searcharxiv⌕ Search

arXiv subjects

Shreya Shah

Publications and source records attributed to Shreya Shah.

10 recordsLinked to original sources

Benefits and Barriers to Human-AI Teaming for Crisis Counseling

Background: AI systems are increasingly deployed to support mental health clinicians, yet real-world evidence remains limited. In crisis counseling, where clients often seek human connection, understanding whether AI augmentation preserves or undermines meaningful human engagement is essential. Methods: In a mixed-methods study of a counselor-facing AI copilot on a WhatsApp-based crisis hotline in India (Mar.-Oct. 2025), counselors could request, edit, and send AI messages. We analyzed 20 interviews, 8 shadowing sessions, 56 survey responses, and edits in 27,778 copilot messages from 108 counselors via thematic analysis (Technology Acceptance Model 2, plus inductive coding of emergent factors), and survey statistics. Results: Counselors reported high copilot acceptance (survey): 98.2% found it helpful; 89.3% likely to continue use; 80.4% reported efficiency gains. Interviews highlighted sentence scaffolding, empathetic phrasing, stress reduction, catching missed details, and efficiency. Barriers (survey) included repetitive outputs (23.2%), lack of context (17.9%), and authenticity concerns (10.7%). While 69.6% (survey) reported skill enhancement, interviews surfaced skill atrophy concerns. Counselors frequently accepted copilot suggestions, editing to remove questions, adjust tone, or personalize style. Conclusion: In one of the largest real-world deployments of a counselor-facing mental health AI copilot, counselors reported reduced cognitive load, empathetic communication, and greater efficiency, alongside concerns about skill atrophy, authenticity, and over-reliance. In settings like crisis counseling where clients seek human connection, AI tools must balance operational support with preserving clinical skill and genuine therapeutic engagement. Designs soliciting approval or editing of complete drafts may overweight efficiency at the expense of sustained human engagement.

cs.HC↗

Beyond a Single Sentiment Score: Topical and Affective Dimensions of Emoji Meaning on Twitter

Emoji are widely treated as compact affective tokens whose meaning is summarised by a single sentiment score. We argue that this picture is incomplete. From a corpus of 16{,}575 English tweets and a per-emoji metric we call \emph{stability delta} (within-class minus between-class cosine similarity of sentence embeddings), we show that emojis used in tightly clustered topical contexts also exhibit \emph{more} polarised crowd-sourced sentiment, not less. Validated against the Kralj Novak et al.\ (2015) sentiment lexicon, the central correlation is Spearman $ρ= +0.361$ (Holm $p = 0.015$, 95\% bootstrap CI $[+0.10, +0.59]$); against the independent Ferré et al.\ (2025) Spanish 13-emotion ratings, $ρ= +0.509$ ($p = 0.0003$). The result survives permutation testing, BCa bootstrap, leave-one-out jackknife, partial correlation controlling for emoji frequency, and Holm correction across a four-test family. Topical coherence and affective dispersion are distinct dimensions of emoji meaning. As a downstream check we report leakage-free five-fold cross-validation results for sarcasm classification on the Emoji Sarcasm Dataset; all three baselines (a gated multimodal architecture, fine-tuned BERTweet features, and fine-tuned RoBERTa features) score above 0.97 macro F1, indicating the dataset has saturating emoji-label cues. We frame the classifier as supporting evidence rather than the contribution.

cs.CL↗

CrisisBench: an evaluation suite for crisis counseling AI

Background: Crisis helplines are essential to mental health systems but are often understaffed and underfunded. LLMs could support counselors, but existing benchmarks do not capture the escalation, risk assessment, and de-escalation central to crisis conversations, while privacy restrictions limit use of real transcripts. Methods: We present CrisisBench, a privacy-preserving suite for single- and multi-turn crisis counseling evaluation. It includes 1,085 single-turn tasks from 200 de-identified conversations and 1,185 patient LLMs derived from hotline metadata. Patient models and a 23-item LLM judge were calibrated against expert counselors. Results: We evaluated five GPT models plus draft-then-edit and retrieval-augmented generation. All systems underperformed human counselors on excessive empathy and questioning in single-turn evaluation. The rubric achieved human-human Krippendorff's $α$ = 0.51 and LLM-human $α$ = 0.54 (Pearson r = 0.80). In multi-turn evaluation, retrieval-augmented generation scored highest overall (0.961), with all systems within 0.058. Microskills and style/tone were near ceiling, while productivity and safety/risk accounted for most between-system variation. Counselors rated patient LLMs realistic (86%) and useful (92%). Conclusion: CrisisBench enables scalable, privacy-preserving evaluation of crisis counseling. Current systems perform strongly on core structural competencies but still diverge from experts on important response-level failure modes. We release the rubrics, judge prompts, and evaluation tasks.

cs.CL↗

"Are you an AI?" Analyzing Client Suspicion of AI Use in Crisis Counseling

As artificial intelligence (AI) tools get increasingly deployed for mental healthcare, public trust in these systems remains uncertain. It is unclear how clients perceive AI involvement in counseling interactions, particularly in moments of crisis that require empathy and connection. To address this gap, we analyzed 75,777 crisis counseling conversations from a human-staffed WhatsApp helpline in India to characterize how often clients suspected they were speaking to AI, what triggered those doubts, and how counselors responded. Though no conversations actually involved AI assistance, the proportion of conversations where clients suspected AI use increased from 0.8% in June 2024 to 2.6% in March 2025. Within suspicious conversations, 21.5% of clients stated an explicit preference for humans. Client suspicion primarily arose in the first half of messages (68.3%), and when counselors offered reassurance (e.g. 'I assure you; this is not ai!'), clients continued to press or ended the conversation 17.6% of the time. As AI tools get increasingly integrated into counselor workflows, understanding these dynamics is essential for designing AI systems that preserve the therapeutic relationship between counselors and clients.

cs.HC↗

DrP: Meta's Efficient Investigations Platform at Scale

Investigations are a significant step in the operational workflows for large scale systems across multiple domains such as services, data, AI/ML, mobile. Investigation processes followed by on-call engineers are often manual or rely on ad-hoc scripts. This leads to inefficient investigations resulting in increased time to mitigate and isolate failures/SLO violations. It also contributes to on-call toil and poor productivity leading to multiple hours/days spent in triaging/debugging incidents. In this paper, we present DrP, an end-to-end framework and system to automate investigations that reduces the mean time to resolve incidents (MTTR) and reduces on-call toil. DrP consists of an expressive and flexible SDK to author investigation playbooks in code (called analyzers), a scalable backend system to execute these automated playbooks, plug-ins to integrate playbooks into mainstream workflows such as alerts and incident management tools, and a post-processing system to take actions on investigations including mitigation steps. We have implemented and deployed DrP at large scale at Meta covering 300+ teams, 2000+ analyzers, across a large set of use cases across domains such as services, core infrastructure, AI/ML, hardware, mobile. DrP has been running in production for the past 5 years and executes 50K automated analyses per day. Overall, our results and experience show that DrP has been able to reduce average MTTR by 20 percent at large scale (with over 80 percent for some teams) and has significantly improved on-call productivity.

cs.SE↗

MedHELM: Holistic Evaluation of Large Language Models for Medical Tasks

While large language models (LLMs) achieve near-perfect scores on medical licensing exams, these evaluations inadequately reflect the complexity and diversity of real-world clinical practice. We introduce MedHELM, an extensible evaluation framework for assessing LLM performance for medical tasks with three key contributions. First, a clinician-validated taxonomy spanning 5 categories, 22 subcategories, and 121 tasks developed with 29 clinicians. Second, a comprehensive benchmark suite comprising 35 benchmarks (17 existing, 18 newly formulated) providing complete coverage of all categories and subcategories in the taxonomy. Third, a systematic comparison of LLMs with improved evaluation methods (using an LLM-jury) and a cost-performance analysis. Evaluation of 9 frontier LLMs, using the 35 benchmarks, revealed significant performance variation. Advanced reasoning models (DeepSeek R1: 66% win-rate; o3-mini: 64% win-rate) demonstrated superior performance, though Claude 3.5 Sonnet achieved comparable results at 40% lower estimated computational cost. On a normalized accuracy scale (0-1), most models performed strongly in Clinical Note Generation (0.73-0.85) and Patient Communication & Education (0.78-0.83), moderately in Medical Research Assistance (0.65-0.75), and generally lower in Clinical Decision Support (0.56-0.72) and Administration & Workflow (0.53-0.63). Our LLM-jury evaluation method achieved good agreement with clinician ratings (ICC = 0.47), surpassing both average clinician-clinician agreement (ICC = 0.43) and automated baselines including ROUGE-L (0.36) and BERTScore-F1 (0.44). Claude 3.5 Sonnet achieved comparable performance to top models at lower estimated cost. These findings highlight the importance of real-world, task-specific evaluation for medical use of LLMs and provides an open source framework to enable this.

cs.CL↗

Epilepsy and its driving forces: understanding the forces behind epileptical pathogenisis

Epilepsy is a neurological disorder characterized by seizures and epileptic events intertwined with religious and personal beliefs since prehistoric times. This review paper explores the historical context and challenges in defining epilepsy. The formal definition was established twenty years ago, and the multifaceted causes of this neurological disorder. It aims to pave the way for personalised therapeutic strategies, research advancements, and informed public health planning to enhance the lives of those affected by this complex neurological condition. In addition, this review paper focuses on the mechanisms and etiologies of epileptogenesis, categorizing them by mechanisms and the underlying causes of the disorder. The review paper provides a brief overview of the current state of the art in the diagnosis, diagnosis, treatment, and treatment of epileptiform seizures.

q-bio.NC↗

Extreme mass-ratio inspirals into black holes surrounded by scalar clouds

We study extreme mass-ratio binary systems in which a stellar mass compact object spirals into a supermassive black hole surrounded by a scalar cloud. Scalar clouds can form through superradiant instabilities of massive scalar fields around spinning black holes and can also serve as a proxy for dark matter halos. Our framework is fully relativistic and assumes that the impact of the cloud on the geometry can be treated perturbatively. As a proof of concept, here we consider a point particle in circular, equatorial motion around a non-spinning black hole surrounded either by a spherically symmetric or a dipolar non-axisymmetric scalar cloud, but the framework can in principle be generalized to generic black hole spins and scalar cloud geometries. We compute the leading-order power lost by the point particle due to scalar radiation and show that, in some regimes, it can dominate over gravitational-wave emission. We confirm the presence of striking signatures due to the presence of a scalar cloud that had been predicted using Newtonian approximations, such as resonances that can give rise to sinking and floating orbits, as well as "sharp features" in the power lost by the particle at given orbital radii. Finally, for a spherically symmetric scalar cloud, we also compute the leading-order corrections to the black-hole geometry and to the gravitational-wave energy flux, focusing only on axial metric perturbations for the latter. We find that, for non-compact clouds, the corrections to the (axial) gravitational-wave fluxes at high frequencies can be understood in terms of a gravitational-redshift effect, in agreement with previous works.

gr-qc↗

Observational constraints on spinning, relativistic Bose-Einstein condensate stars

Bose-Einstein condensates (BECs) have been proposed as candidate states of matter for the interior of neutron stars. Specifically, Chavanis and Harko obtained the mass-radius relation for a BEC star and proposed that the recently discovered neutron stars with masses around 2$M_\odot$ are BEC stars. They employed a barotropic equation of state (EOS), with one free parameter, that was first found by Colpi, Wasserman, and Shapiro (CSW), to describe them and derive stable equilibrium configurations of spinning BEC stars in General Relativity. In this work we show that while it is true that BECs allow for compact object masses as heavy as the heaviest observed ones, such stars cannot simultaneously have radii that are small enough to be consistent with the latest observations, in spite of the flexibility available in the EOS in the form of the free parameter. In fact, our conclusion applies to any spinning relativistic boson star that obeys the CSW EOS.

astro-ph.HE↗

Social Network for Smart Devices using Embedded Ethernet

Embedded Ethernet is nothing but a microcontroller which is able to communicate with the network. A design of AVR controller-based embedded Ethernet interface is presented. In the design, an existing SPI serial device can be converted into a network interface peripheral to obtain compatibility with the network. By typing the IP-address of LAN on the web browser, the user gets a web page on screen; this page contains all the information about the status of the devices. The user can also control the devices interfaced to the web server by pressing buttons provided in the web page. This creates a network for easy communication among the devices.

cs.NI↗