Hasil untuk "math.QA"

E Kremmel, T Giaddui, J Keller et al.

Purpose:This study compares gamma passing rates for a cohort of similar intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT) quality assurance (QA) plans to determine the equivalence of the patient specific QA plan delivery. The passing criterion is 90% gamma index with 3% dose difference (DD), 3mm distance‐to‐agreement (DTA) and a dose threshold of 10%.Methods:Gamma passing results of previously treated IMRT QA plans, delivered to Sun Nuclear MapCheck phantom, and VMAT QA plans, delivered to ScandiDos Delta4 phantom, are organized by anatomical site and treatment machine. Both Varian and Elekta machines are included. Pairs of IMRT and VMAT plans are matched based on site, machine, and PTV volume to ensure similar plan cohorts. A two‐tailed t‐test analysis of the data with an alpha of 0.05 determines if there exists a statistically significant difference. Power was calculated to detect a difference of 4%; all data sets were strong with above a 0.8 power.Results:The first data set consisting of 20 matched pairs of prostate plans was statistically insignificant (p‐value=0.90, Power=0.99). The 14 matched pairs set of head and neck plans has a statistically significant Result (p‐value=0.028, Power=0.88). The head and neck IMRT gamma indexes have a mean of 93.1% and range of 82%‐100% while the VMAT gamma indexes have a mean of 96.7% and range of 92%‐100%. The two combined data sets of matched plans had a statistically insignificant Result (p‐value=0.073, Power=0.99).Conclusion:Overall, IMRT and VMAT have equivalent passing rates when comparing the gamma analysis using a passing criterion of 3% DD and 3mm DTA. When separated by site, prostate IMRT and VMAT plans have equivalent passing rates while head and neck plans have a statistically significant variation of passing rates. The passing rates for the two modalities are independent of delivery machine for matched PTV target volumes.

N. Groom, E. Wilson, C. Faivre-Finn et al.

N. Khandke

W.J. Tilstone

Kirtan Brijeshbhai Soni, Krish Rupapara, Arpit Rana et al.

Accelerating research in renewable energy policy is critical for addressing climate change and enabling informed decision-making. Question answering (QA) over public policy documents presents unique challenges due to their legal structure, conditional dependencies, and domain-specific vocabulary. In this paper, we introduce EvalQAG, a framework for generating high-quality QA pairs from renewable energy policy documents. EvalQAG combines structured prompts, retrieval-augmented inputs, and multi-stage evaluation using large language models (LLMs) to support accurate and diverse QA generation. Using this framework, we construct REPolicyQA, a domain-specific QA dataset comprising approximately 160,000 QA pairs from over 1,000 U.S. renewable energy policy documents. The dataset covers five policy-relevant question types: Yes/No, Yes/No with Conditions, Factual, Legal Obligation, and Descriptive, which capture a wide range of reasoning patterns grounded in regulatory texts. We evaluate multiple QA models and uncover significant performance gaps, particularly in legal reasoning and conditional inference, highlighting major shortcomings in current systems. Our results establish EvalQAG as a generalizable QA generation pipeline for policy texts and position REPolicyQA as a new benchmark for advancing QA research in policy and regulatory domains. We believe this work can foster impactful research in the renewable energy sector, particularly by enabling more robust and explainable QA systems for legal and condition-heavy regulatory documents.

Phanindra Sai Boyapati, Sudhakar Allam

Independent Researcher, Sanjay Kumar Das

Providers of retirement plans are facing increasing pressure to modernize loan and distribution processing in 401(k) and other retirement systems to improve efficiency, accuracy, and compliance. Legacy processing environments often mean manual processes and fragmented legacy systems, which mean long turnaround times and a high error rate. This paper proposes a Quality Assurance (QA) automation approach to update these critical processes. We design and implement an automated test and processing framework that leverages behavior-driven development and state-of-the-art test automation tools to enable straight-through processing and comprehensive verification of loan and distribution transactions. The method was applied to real retirement system environments with much shorter processing cycles and quasi-elimination of processing errors. Results demonstrate that automation-driven modernization can provide 50% faster processing, improved compliance with regulatory requirements, and improved participant satisfaction. The findings of the research have significant implications for the modernization of financial technology, illustrating how QA automation can bridge the gap between legacy retirement platforms and the demands of contemporary financial services, while ensuring robust compliance and reliability.

James S. Michaelson

Redazione

Redazione

Redazione

Lutz Cauers, Max Häge

P. Purser

C.E. WILLS

C.E. WILLS

Anna Łyczkowska