arXiv subjects

Mohit Jain

Publications and source records attributed to Mohit Jain.

At least 19 recordsLinked to original sources

arXiv2026

MeDxAgent: Multi-Agent Consultation for Interactive Medical Diagnosis

Large language models (LLMs) are increasingly used for health-related decision support. Yet most evaluations treat diagnosis as a single-shot task with complete information provided upfront, often as a multiple-choice selection. This diverges from clinical practice, where diagnosis is interactive and open-ended, involving sequential hypothesis refinement through targeted questioning. We address this gap. We build MeDxBench, a large-scale benchmark of 4,421 clinical cases across 20 specialties. We further propose MeDxAgent, a multi-agent consultation system for interactive diagnosis, and systematically study its prompt-, flow- and agent-level design choices. MeDxAgent achieves a 10.3% accuracy gain over the baseline on MeDxBench, closing 52.3% of the gap to a full-information oracle. We find that specific design choices: collecting demographics first, passing summarized dialogue for diagnosis, and feeding candidate diagnoses for targeted questioning, improve accuracy, mirroring how physicians reason, though their effect emerges fully only in combination. Code and dataset will be released upon publication.

cs.MA↗

arXiv2026

Designing Medical Chatbots where Accuracy and Acceptability are in Conflict: An Exploratory, Vignette-based Study in Urban India

When medical chatbots provide advice that conflicts with users' lived care experiences, users are left to interpret, negotiate, and evaluate the legitimacy of that guidance. In India, the widespread overuse of antibiotics, antidiarrheals, and injections has shifted patient expectations away from the guideline-aligned advice that chatbots are trained to provide. We present a mixed-methods, vignette-based study with 200 urban Indian adults examining preferences for and against guideline-aligned, norm-divergent advice in chatbot transcripts. We find that a majority of users reject such advice, drawing on diverse rationales grounded in their lived expectations. Through the design and introduction of context-aware nudges, we support expectation alignment that shifts preferences towards transcripts containing guideline-aligned advice. In doing so, we surface key tensions in the equitable design of medical chatbots in the Global South.

cs.HC↗

arXiv2025

Editing with AI: How Doctors Refine LLM-Generated Answers to Patient Queries

Patients frequently seek information during their medical journeys, but the rising volume of digital patient messages has strained healthcare systems. Large language models (LLMs) offer promise in generating draft responses for clinicians, yet how physicians refine these drafts remains underexplored. We present a mixed-methods study with nine ophthalmologists answering 144 cataract surgery questions across three conditions: writing from scratch, directly editing LLM drafts, and instruction-based indirect editing. Our quantitative and qualitative analyses reveal that while LLM outputs were generally accurate, occasional errors and automation bias revealed the need for human oversight. Contextualization--adapting generic answers to local practices and patient expectations--emerged as a dominant form of editing. Editing workflows revealed trade-offs: indirect editing reduced effort but introduced errors, while direct editing ensured precision but with higher workload. We conclude with design and policy implications for building safe, scalable LLM-assisted clinical communication systems.

cs.HC↗

arXiv2025

CataractCompDetect: Intraoperative Complication Detection in Cataract Surgery

Cataract surgery is one of the most commonly performed surgeries worldwide, yet intraoperative complications such as iris prolapse, posterior capsule rupture (PCR), and vitreous loss remain major causes of adverse outcomes. Automated detection of such events could enable early warning systems and objective training feedback. In this work, we propose CataractCompDetect, a complication detection framework that combines phase-aware localization, SAM 2-based tracking, complication-specific risk scoring, and vision-language reasoning for final classification. To validate CataractCompDetect, we curate CataComp, the first cataract surgery video dataset annotated for intraoperative complications, comprising 53 surgeries, including 23 with clinical complications. On CataComp, CataractCompDetect achieves an average F1 score of 70.63%, with per-complication performance of 81.8% (Iris Prolapse), 60.87% (PCR), and 69.23% (Vitreous Loss). These results highlight the value of combining structured surgical priors with vision-language reasoning for recognizing rare but high-impact intraoperative events. Our dataset and code will be publicly released upon acceptance.

cs.CV↗

arXiv2025

Closing the Performance Gap Between AI and Radiologists in Chest X-Ray Reporting

AI-assisted report generation offers the opportunity to reduce radiologists' workload stemming from expanded screening guidelines, complex cases and workforce shortages, while maintaining diagnostic accuracy. In addition to describing pathological findings in chest X-ray reports, interpreting lines and tubes (L&T) is demanding and repetitive for radiologists, especially with high patient volumes. We introduce MAIRA-X, a clinically evaluated multimodal AI model for longitudinal chest X-ray (CXR) report generation, that encompasses both clinical findings and L&T reporting. Developed using a large-scale, multi-site, longitudinal dataset of 3.1 million studies (comprising 6 million images from 806k patients) from Mayo Clinic, MAIRA-X was evaluated on three holdout datasets and the public MIMIC-CXR dataset, where it significantly improved AI-generated reports over the state of the art on lexical quality, clinical correctness, and L&T-related elements. A novel L&T-specific metrics framework was developed to assess accuracy in reporting attributes such as type, longitudinal change and placement. A first-of-its-kind retrospective user evaluation study was conducted with nine radiologists of varying experience, who blindly reviewed 600 studies from distinct subjects. The user study found comparable rates of critical errors (3.0% for original vs. 4.6% for AI-generated reports) and a similar rate of acceptable sentences (97.8% for original vs. 97.4% for AI-generated reports), marking a significant improvement over prior user studies with larger gaps and higher error rates. Our results suggest that MAIRA-X can effectively assist radiologists, particularly in high-volume clinical settings.

cs.CL↗

arXiv2025

Trade-offs in Social-Norm Framings for Health Chatbots: Balancing Trust and Preference

AI-driven chatbots are increasingly being used to support community health workers (CHWs) in developing regions. Yet little is known about how cultural frameworks in chatbot design shape trust in collectivist contexts where decisions are rarely made in isolation. This paper examines how CHWs in rural India responded to chatbot-interfaces that delivered identical health content but varied in one specific cultural lever: social norms. Through a mixed-methods study with 61 ASHAs who compared four normative framings: neutral, descriptive, narrative identity, and injunctive authority, we (1) analyze how framings influence preferences and trust and (2) compare effects in low- and high-ambiguity scenarios. The results show that narrative framings were most preferred but encouraged overreliance, while authority framings were least preferred yet supported calibrated trust. We conclude with design recommendations for dynamic framing strategies that adapt to context and argue for calibrated trust--following correct advice and resisting incorrect advice--as a critical evaluation metric for safe, culturally-grounded AI.

cs.HC↗

arXiv2025

Designing Culturally Aligned AI Systems For Social Good in Non-Western Contexts

AI technologies are increasingly deployed in high-stakes domains such as education, healthcare, law, and agriculture to address complex challenges in non-Western contexts. This paper examines eight real-world deployments spanning seven countries and 18 languages, combining 17 interviews with AI developers and domain experts with secondary research. Our findings identify six cross-cutting factors - Language, Institution, Safety, Task, End-User Demography, and Domain - that structured how systems were designed and deployed. These factors were shaped by Sociocultural (diversity, practices), Institutional (resources, policies), and Technological (capabilities, limits) influences. We find that building effective AI systems required extensive collaboration between AI developers and domain experts, with human resources proving more critical to achieving safe and effective outcomes in high-stakes domains than technological expertise alone. Additionally, we present 12 guidelines synthesizing these dynamics for designing AI for social good systems that are culturally grounded, equitable, and responsive to the needs of non-Western contexts.

cs.HC↗

arXiv2024

Phase-Informed Tool Segmentation for Manual Small-Incision Cataract Surgery

Cataract surgery is the most common surgical procedure globally, with a disproportionately higher burden in developing countries. While automated surgical video analysis has been explored in general surgery, its application to ophthalmic procedures remains limited. Existing works primarily focus on Phaco cataract surgery, an expensive technique not accessible in regions where cataract treatment is most needed. In contrast, Manual Small-Incision Cataract Surgery (MSICS) is the preferred low-cost, faster alternative in high-volume settings and for challenging cases. However, no dataset exists for MSICS. To address this gap, we introduce Sankara-MSICS, the first comprehensive dataset containing 53 surgical videos annotated for 18 surgical phases and 3,527 frames with 13 surgical tools at the pixel level. We benchmark this dataset on state-of-the-art models and present ToolSeg, a novel framework that enhances tool segmentation by introducing a phase-conditional decoder and a simple yet effective semi-supervised setup leveraging pseudo-labels from foundation models. Our approach significantly improves segmentation performance, achieving a $23.77\%$ to $38.10\%$ increase in mean Dice scores, with a notable boost for tools that are less prevalent and small. Furthermore, we demonstrate that ToolSeg generalizes to other surgical settings, showcasing its effectiveness on the CaDIS dataset.

cs.CV↗

arXiv2024

HEALTH-PARIKSHA: Assessing RAG Models for Health Chatbots in Real-World Multilingual Settings

Assessing the capabilities and limitations of large language models (LLMs) has garnered significant interest, yet the evaluation of multiple models in real-world scenarios remains rare. Multilingual evaluation often relies on translated benchmarks, which typically do not capture linguistic and cultural nuances present in the source language. This study provides an extensive assessment of 24 LLMs on real world data collected from Indian patients interacting with a medical chatbot in Indian English and 4 other Indic languages. We employ a uniform Retrieval Augmented Generation framework to generate responses, which are evaluated using both automated techniques and human evaluators on four specific metrics relevant to our application. We find that models vary significantly in their performance and that instruction tuned Indic models do not always perform well on Indic language queries. Further, we empirically show that factual correctness is generally lower for responses to Indic queries compared to English queries. Finally, our qualitative work shows that code-mixed and culturally relevant queries in our dataset pose challenges to evaluated models.

cs.CL↗

arXiv2024

Learnings from a Large-Scale Deployment of an LLM-Powered Expert-in-the-Loop Healthcare Chatbot

Large Language Models (LLMs) are widely used in healthcare, but limitations like hallucinations, incomplete information, and bias hinder their reliability. To address these, researchers released the Build Your Own expert Bot (BYOeB) platform, enabling developers to create LLM-powered chatbots with integrated expert verification. CataractBot, its first implementation, provides expert-verified responses to cataract surgery questions. A pilot evaluation showed its potential; however the study had a small sample size and was primarily qualitative. In this work, we conducted a large-scale 24-week deployment of CataractBot involving 318 patients and attendants who sent 1,992 messages, with 91.71% of responses verified by seven experts. Analysis of interaction logs revealed that medical questions significantly outnumbered logistical ones, hallucinations were negligible, and experts rated 84.52% of medical answers as accurate. As the knowledge base expanded with expert corrections, system performance improved by 19.02%, reducing expert workload. These insights guide the design of future LLM-powered chatbots.

cs.HC↗

arXiv2024

ASHABot: An LLM-Powered Chatbot to Support the Informational Needs of Community Health Workers

Community health workers (CHWs) provide last-mile healthcare services but face challenges due to limited medical knowledge and training. This paper describes the design, deployment, and evaluation of ASHABot, an LLM-powered, experts-in-the-loop, WhatsApp-based chatbot to address the information needs of CHWs in India. Through interviews with CHWs and their supervisors and log analysis, we examine factors affecting their engagement with ASHABot, and ASHABot's role in addressing CHWs' informational needs. We found that ASHABot provided a private channel for CHWs to ask rudimentary and sensitive questions they hesitated to ask supervisors. CHWs trusted the information they received on ASHABot and treated it as an authoritative resource. CHWs' supervisors expanded their knowledge by contributing answers to questions ASHABot failed to answer, but were concerned about demands on their workload and increased accountability. We emphasize positioning LLMs as supplemental fallible resources within the community healthcare ecosystem, instead of as replacements for supervisor support.

cs.HC↗

arXiv2024

A Deadline-Aware Scheduler for Smart Factory using WiFi 6

A key strategy for making production in factories more efficient is to collect data about the functioning of machines, and dynamically adapt their working. Such smart factories have data packets with a mix of stringent and non-stringent deadlines with varying levels of importance that need to be delivered via a wireless network. However, the scheduling of packets in the wireless network is crucial to satisfy the deadlines. In this work, we propose a technique of utilizing IEEE 802.11ax, popularly known as WiFi 6, for such applications. IEEE 802.11ax has a few unique characteristics, such as specific configurations of dividing the channels into resource units (RU) for packet transmission and synchronized parallel transmissions. We model the problem of scheduling packets by assigning profit to each packet and then maximizing the sum of profits. We first show that this problem is strongly NP-Hard, and then propose an approximation algorithm with a 12-approximate algorithm. Our approximation algorithm uses a variant of local search to associate the right RU configuration to each packet and identify the duration of each parallel transmission. Finally, we extensively simulate different scenarios to show that our algorithm works better than other benchmarks.

cs.NI↗

arXiv2024

CataractBot: An LLM-Powered Expert-in-the-Loop Chatbot for Cataract Patients

The healthcare landscape is evolving, with patients seeking reliable information about their health conditions and available treatment options. Despite the abundance of information sources, the digital age overwhelms individuals with excess, often inaccurate information. Patients primarily trust medical professionals, highlighting the need for expert-endorsed health information. However, increased patient loads on experts has led to reduced communication time, impacting information sharing. To address this gap, we developed CataractBot. CataractBot answers cataract surgery related questions instantly using an LLM to query a curated knowledge base, and provides expert-verified responses asynchronously. It has multimodal and multilingual capabilities. In an in-the-wild deployment study with 49 patients and attendants, 4 doctors, and 2 patient coordinators, CataractBot demonstrated potential, providing anytime accessibility, saving time, accommodating diverse literacy levels, alleviating power differences, and adding a privacy layer between patients and doctors. Users reported that their trust in the system was established through expert verification. Broadly, our results could inform future work on expert-mediated LLM bots.

cs.HC↗

arXiv2023

The Art of Embedding Fusion: Optimizing Hate Speech Detection

Hate speech detection is a challenging natural language processing task that requires capturing linguistic and contextual nuances. Pre-trained language models (PLMs) offer rich semantic representations of text that can improve this task. However there is still limited knowledge about ways to effectively combine representations across PLMs and leverage their complementary strengths. In this work, we shed light on various combination techniques for several PLMs and comprehensively analyze their effectiveness. Our findings show that combining embeddings leads to slight improvements but at a high computational cost and the choice of combination has marginal effect on the final outcome. We also make our codebase public at https://github.com/aflah02/The-Art-of-Embedding-Fusion-Optimizing-Hate-Speech-Detection .

cs.CL↗

arXiv2023

PwR: Exploring the Role of Representations in Conversational Programming

Large Language Models (LLMs) have revolutionized programming and software engineering. AI programming assistants such as GitHub Copilot X enable conversational programming, narrowing the gap between human intent and code generation. However, prior literature has identified a key challenge--there is a gap between user's mental model of the system's understanding after a sequence of natural language utterances, and the AI system's actual understanding. To address this, we introduce Programming with Representations (PwR), an approach that uses representations to convey the system's understanding back to the user in natural language. We conducted an in-lab task-centered study with 14 users of varying programming proficiency and found that representations significantly improve understandability, and instilled a sense of agency among our participants. Expert programmers use them for verification, while intermediate programmers benefit from confirmation. Natural language-based development with LLMs, coupled with representations, promises to transform software development, making it more accessible and efficient.

cs.HC↗

arXiv2023

"Can't Take the Pressure?": Examining the Challenges of Blood Pressure Estimation via Pulse Wave Analysis

The use of observed wearable sensor data (e.g., photoplethysmograms [PPG]) to infer health measures (e.g., glucose level or blood pressure) is a very active area of research. Such technology can have a significant impact on health screening, chronic disease management and remote monitoring. A common approach is to collect sensor data and corresponding labels from a clinical grade device (e.g., blood pressure cuff), and train deep learning models to map one to the other. Although well intentioned, this approach often ignores a principled analysis of whether the input sensor data has enough information to predict the desired metric. We analyze the task of predicting blood pressure from PPG pulse wave analysis. Our review of the prior work reveals that many papers fall prey data leakage, and unrealistic constraints on the task and the preprocessing steps. We propose a set of tools to help determine if the input signal in question (e.g., PPG) is indeed a good predictor of the desired label (e.g., blood pressure). Using our proposed tools, we have found that blood pressure prediction using PPG has a high multi-valued mapping factor of 33.2% and low mutual information of 9.8%. In comparison, heart rate prediction using PPG, a well-established task, has a very low multi-valued mapping factor of 0.75% and high mutual information of 87.7%. We argue that these results provide a more realistic representation of the current progress towards to goal of wearable blood pressure measurement via PPG pulse wave analysis.

eess.SP↗

arXiv2023

Understanding Journalists' Workflows in News Curation

With the increasing dominance of the internet as a source of news consumption, there has been a rise in the production and popularity of email newsletters compiled by individual journalists. However, there is little research on the processes of aggregation, and how these differ between expert journalists and trained machines. In this paper, we interviewed journalists who curate newsletters from around the world. Through an in-depth understanding of journalists' workflows, our findings lay out the role of their prior experience in the value they bring into the curation process, their use of algorithms in finding stories for their newsletter, and their internalization of their readers' interests and the context they are curating for. While identifying the role of human expertise, we highlight the importance of hybrid curation and provide design insights on how technology can support the work of these experts.

cs.CY↗

arXiv2023

Visual Reinforcement Learning with Self-Supervised 3D Representations

A prominent approach to visual Reinforcement Learning (RL) is to learn an internal state representation using self-supervised methods, which has the potential benefit of improved sample-efficiency and generalization through additional learning signal and inductive biases. However, while the real world is inherently 3D, prior efforts have largely been focused on leveraging 2D computer vision techniques as auxiliary self-supervision. In this work, we present a unified framework for self-supervised learning of 3D representations for motor control. Our proposed framework consists of two phases: a pretraining phase where a deep voxel-based 3D autoencoder is pretrained on a large object-centric dataset, and a finetuning phase where the representation is jointly finetuned together with RL on in-domain data. We empirically show that our method enjoys improved sample efficiency in simulated manipulation tasks compared to 2D representation learning methods. Additionally, our learned policies transfer zero-shot to a real robot setup with only approximate geometric correspondence, and successfully solve motor control tasks that involve grasping and lifting from a single, uncalibrated RGB camera. Code and videos are available at https://yanjieze.com/3d4rl/ .

cs.LG↗