collaborators

5 papers

cs.LG2026

Revalidation Beats Stateful Routing for Scientific Surrogates Under Distribution Shift

Harshil Lodhiya

Surrogate models are often chosen during development and then left in place as new measurements arrive. That practice becomes risky when noise, input support, or physical parameter…

cs.AI2026

Clinical Knowledge Graphs for Chest X-Ray Device Reasoning

Harshil Lodhiya

Chest radiographs are routinely used to verify the position of catheters, tubes, and other support devices. Existing image models often return labels or segmentations, while report…

cs.LG2026

ER-KANs: Efficient and Robust Kolmogorov-Arnold Networks for Data-Scarce Scientific Machine Learning

Harshil Lodhiya

The efficient-KAN literature---covering Chebyshev, wavelet, and radial-basis-function variants of the original Kolmogorov-Arnold Network---has been benchmarked almost entirely on c…

cs.AI2026

Agentic Governance and Adversarial Verification for Policy-Constrained LLM Healthcare Appeal Generation

Harshil Lodhiya, Alex McManus, Reese Walker

Claim denial management costs U.S. healthcare approximately $260 billion annually in administrative overhead. Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG)…

eess.IV2026

Uncertainty-Aware Compositional Localization and Placement Assessment of Catheters and Tubes in Chest X-Rays

Harshil Lodhiya

Assessing catheter and tube placement on chest X-rays is safety-critical yet tedious and error-prone. Current deep learning methods either classify placement globally -- losing tra…