arXiv subjects

Aman Dalmia

Publications and source records attributed to Aman Dalmia.

8 recordsLinked to original sources

arXiv2026

Auditable Emergency Triage for Maternal and Newborn Care in India

At Noora Health, our nurses answer more than 50,000 medical queries per month on our WhatsApp-based service that provides caregivers with on-demand support. Their most time-critical task is emergency triage: deciding which queries need immediate in-person attention. To support them, we built a system that uses a large language model (LLM) to classify whether a message is an emergency and provide a rationale for interpretability. But the system was opaque: analyzing mistakes meant reading reasoning chains for each message, which is infeasible at our scale. Prompt changes meant re-running a full evaluation to prevent regressions, which was both costly and operationally challenging. Clinicians follow a decision tree to make this call, but it was never documented or passed to the model, which relied on a flat list of danger signs. To address these issues, we decomposed triage into two steps: an LLM extracts canonical symptoms and patient context from the query using a clinician-authored vocabulary, and a deterministic rule engine captures the scenarios that indicate an emergency. We show that the new system raised recall from 0.565 to 0.810 and F1 from 0.606 to 0.702, with structured rules driving most of the accuracy gains while the decomposition provides auditability: clinical experts can inspect each stage of the new system to see whether the query was mistranslated, symptoms were incorrectly extracted, patient context was wrongly inferred, or the necessary rules were missing. They can add new rules independently without causing regressions and avoid running costly evaluations. Since deployment, the new system has triaged 152,421 patient queries and flagged 28,535 (18.7%) as emergencies. The over-escalation rate has been 17.8%, without any increase in missed emergencies. Clinicians have also added 48 new rules since deployment, evidence of the faster correction loop we set out to build.

cs.CL↗

arXiv2026

FormBharo: Designing and Evaluating a Voice Agent for Conversational Form Filling in Rural India

In India, almost every social benefit starts with a form, yet the people who need these benefits most are often unable to read or write. Reaching them requires a spoken conversation. Today that work falls to frontline health workers who enroll beneficiaries one at a time, a poor use of stretched capacity. We built FormBharo ("fill the form" in Hindi), a voice agent that fills a structured form over a phone call under tight latency and cost budgets by pairing Large Language Models (LLMs) with deterministic, rule-based validation and flow control. It is being piloted with ARMMAN, an NGO running large-scale maternal and child mobile-health programs in India, to enroll low-income, Hindi-speaking mothers in antenatal and postnatal care. To our knowledge, it is the first voice agent piloted to fill an enrollment form for this population. We openly release FormVoiceAgentBench, a benchmark pairing human-recorded Hindi audio with 3,760 multi-turn conversation tests across 960 simulated calls, to evaluate our agent's components (transcription, extraction, reply generation) and end-to-end form completion under real acoustic variations. Form completion drops by up to ~41 points when LLMs receive error-prone real-speech transcripts instead of reference ones. The rule-based controls recover many turn-level extraction errors, helping smaller, cheaper models match or surpass frontier models on form completion. Component performance does not predict end-to-end performance: GPT-5.5 leads turn-level extraction accuracy on reference transcripts (99.8%) but ranks lower on form completion. Since errors both propagate and cancel across the pipeline, the optimal model choice of models emerges only through end-to-end evaluation. Finally, no single model is best across accuracy, cost, and latency at once, so we use a Pareto-based weighted-sum scalarization to select a deployable configuration balancing the three.

cs.CL↗

arXiv2025

Script Gap: Evaluating LLM Triage on Indian Languages in Native vs Romanized Scripts in a Real World Setting

Large Language Models (LLMs) are increasingly deployed in high-stakes clinical applications in India. Speakers of Indian languages frequently communicate using romanized text rather than native scripts, yet existing research rarely quantifies or evaluates this orthographic variation in real world applications. We investigate how romanization impacts the reliability of LLMs in a critical domain: maternal and newborn healthcare triage. We benchmark leading LLMs on a real world dataset of user-generated health queries spanning five Indian languages and Nepali. Our results reveal consistent degradation in performance for romanized messages, with gap reaching up to 24 points across languages and models. We propose and evaluate an Uncertainty-based Selective Routing method to close this script gap. At our partner maternal health organization alone, this gap could cause nearly 2 million excess errors in triage. Our findings highlight a critical safety blind spot in LLM-based health systems: models that appear to understand romanized input may still fail to act on it reliably.

cs.CL↗

arXiv2024

GeoCoder: Solving Geometry Problems by Generating Modular Code through Vision-Language Models

Geometry problem-solving demands advanced reasoning abilities to process multimodal inputs and employ mathematical knowledge effectively. Vision-language models (VLMs) have made significant progress in various multimodal tasks. Yet, they still struggle with geometry problems and are significantly limited by their inability to perform mathematical operations not seen during pre-training, such as calculating the cosine of an arbitrary angle, and by difficulties in correctly applying relevant geometry formulas. To overcome these challenges, we present GeoCoder, which leverages modular code-finetuning to generate and execute code using a predefined geometry function library. By executing the code, we achieve accurate and deterministic calculations, contrasting the stochastic nature of autoregressive token prediction, while the function library minimizes errors in formula usage. We also propose a multimodal retrieval-augmented variant of GeoCoder, named RAG-GeoCoder, which incorporates a non-parametric memory module for retrieving functions from the geometry library, thereby reducing reliance on parametric memory. Our modular code-finetuning approach enhances the geometric reasoning capabilities of VLMs, yielding an average improvement of over 16% across various question complexities on the GeomVerse dataset compared to other finetuning methods.

cs.CL↗

arXiv2024

NurtureNet: A Multi-task Video-based Approach for Newborn Anthropometry

Malnutrition among newborns is a top public health concern in developing countries. Identification and subsequent growth monitoring are key to successful interventions. However, this is challenging in rural communities where health systems tend to be inaccessible and under-equipped, with poor adherence to protocol. Our goal is to equip health workers and public health systems with a solution for contactless newborn anthropometry in the community. We propose NurtureNet, a multi-task model that fuses visual information (a video taken with a low-cost smartphone) with tabular inputs to regress multiple anthropometry estimates including weight, length, head circumference, and chest circumference. We show that visual proxy tasks of segmentation and keypoint prediction further improve performance. We establish the efficacy of the model through several experiments and achieve a relative error of 3.9% and mean absolute error of 114.3 g for weight estimation. Model compression to 15 MB also allows offline deployment to low-cost smartphones.

cs.CV↗

arXiv2021

Impact of data-splits on generalization: Identifying COVID-19 from cough and context

Rapidly scaling screening, testing and quarantine has shown to be an effective strategy to combat the COVID-19 pandemic. We consider the application of deep learning techniques to distinguish individuals with COVID from non-COVID by using data acquirable from a phone. Using cough and context (symptoms and meta-data) represent such a promising approach. Several independent works in this direction have shown promising results. However, none of them report performance across clinically relevant data splits. Specifically, the performance where the development and test sets are split in time (retrospective validation) and across sites (broad validation). Although there is meaningful generalization across these splits the performance significantly varies (up to 0.1 AUC score). In addition, we study the performance of symptomatic and asymptomatic individuals across these three splits. Finally, we show that our model focuses on meaningful features of the input, cough bouts for cough and relevant symptoms for context. The code and checkpoints are available at https://github.com/WadhwaniAI/cough-against-covid

cs.SD↗

arXiv2020

Cough Against COVID: Evidence of COVID-19 Signature in Cough Sounds

Testing capacity for COVID-19 remains a challenge globally due to the lack of adequate supplies, trained personnel, and sample-processing equipment. These problems are even more acute in rural and underdeveloped regions. We demonstrate that solicited-cough sounds collected over a phone, when analysed by our AI model, have statistically significant signal indicative of COVID-19 status (AUC 0.72, t-test,p <0.01,95% CI 0.61-0.83). This holds true for asymptomatic patients as well. Towards this, we collect the largest known(to date) dataset of microbiologically confirmed COVID-19 cough sounds from 3,621 individuals. When used in a triaging step within an overall testing protocol, by enabling risk-stratification of individuals before confirmatory tests, our tool can increase the testing capacity of a healthcare system by 43% at disease prevalence of 5%, without additional supplies, trained personnel, or physical infrastructure

cs.SD↗

arXiv2018

Siamese Neural Networks with Random Forest for detecting duplicate question pairs

Determining whether two given questions are semantically similar is a fairly challenging task given the different structures and forms that the questions can take. In this paper, we use Gated Recurrent Units(GRU) in combination with other highly used machine learning algorithms like Random Forest, Adaboost and SVM for the similarity prediction task on a dataset released by Quora, consisting of about 400k labeled question pairs. We got the best result by using the Siamese adaptation of a Bidirectional GRU with a Random Forest classifier, which landed us among the top 24% in the competition Quora Question Pairs hosted on Kaggle.

cs.CL↗