惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

爱范儿
爱范儿
WordPress大学
WordPress大学
C
Check Point Blog
GbyAI
GbyAI
U
Unit 42
Google DeepMind News
Google DeepMind News
B
Blog RSS Feed
Blog — PlanetScale
Blog — PlanetScale
J
Java Code Geeks
I
InfoQ
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Hugging Face - Blog
Hugging Face - Blog
Vercel News
Vercel News
博客园 - 【当耐特】
美团技术团队
小众软件
小众软件
S
SegmentFault 最新的问题
Jina AI
Jina AI
阮一峰的网络日志
阮一峰的网络日志
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
The Cloudflare Blog
Last Week in AI
Last Week in AI
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
V
Visual Studio Blog

Business News Today: Latest Business News, Finance News

Markets’ dilemma: Trust the bark or wag of oil prices The sector call illusion Bandu’s Blockbusters For April 12, 2026 Mastering Derivatives: Does Lag Impact Effectiveness Of OI? Who Am I? April 12, 2026 Index Outlook: Rising From Dire Straits US Market Outlook: Gaining Strength Bullion Cues: Gold And Silver Futures Face Barrier F&O Tracker: Tentative Shift In Trend F&O Strategy: Buy L&T Put Maruti Suzuki to launch 4 EVs by 2031 India Inc flags surge in cost of packaging raw material, seeks relief measures India-flagged LPG tanker Jag Vikram crosses Strait of Hormuz after US-Iran ceasefire Muted pricing power, rising costs to curb benefits of demand in cement sector: HDFC Securities Iran's new supreme leader Mojtaba Khamenei has severe and disfiguring wounds, sources say No road tax, registration fees for electric vehicles priced up to ₹30 lakh till March 2030: Delhi’s draft EV policy Central Railway to run four special local trains for Ambedkar Jayanti West Asia tensions push up costs for India; further impact hinges on stability: Report ED initiates fresh raids against former Bengal minister Chatterjee in teacher recruitment scam Election Commission reverses Mittal’s DVAC posting, appoints him DGP, TN Armed Police Israel and Lebanon are expected to hold talks. Here’s what to know US, Iran set for peace talks but doubts emerge over Lebanon, sanctions Cotton Association revises output estimates for 2025-26 up at 324 lakh bales of 170 kg each Orbicular gets USFDA’s tentative nod for generic Semaglutide Injection in partnership with Apotex Malls, high-streets in NCR clock 45% rise in leasing of retail spaces in Jan-Mar: C&W FIIs pull ₹28,375 crore in five sessions; domestic buyers cushion fall as indices post best week in months Nifty and Bank Nifty Prediction for the week 13 Apr’26 to 17 Apr’26 by BL GURU Proposed Trump arch in Washington DC includes winged figure, eagles, lions and gold inscriptions 'Ladakh' replaces 'Jammu and Kashmir' in Aadhaar records for UT residents Misri ends US trip with focus on civil nuclear cooperation and LPG exports
Zorgm Pro achieves 98.5% on NEET PG benchmark with source...
cuelive · 2026-06-24 · via Business News Today: Latest Business News, Finance News
Zorgm Pro achieves 98.5% on NEET PG benchmark with source-grounded clinical AI

Zorgm Pro achieves 98.5% on NEET PG benchmark with source-grounded clinical AI

Educational reference answer engine for verified physicians becomes one of the first source-grounded clinical AI systems to surpass the 95% threshold on a publicly verifiable NEET PG benchmark

TL;DR

  • Achieved a 98.54% score on the NEET PG 2025 benchmark using a publicly documented evaluation methodology
  • Every answer is backed by citations from peer-reviewed journals, clinical guidelines, and regulator-approved sources to help reduce AI hallucinations
  • Available free for verified physicians, with clinical-language outputs designed specifically for medical education and practice

MUMBAI, India , June 24, 2026 /PRNewswire/ -- In a major breakthrough for clinical artificial intelligence, the LaennecAI Clinical Research Team announced today that Zorgm Pro , an educational reference answer engine built exclusively for verified physicians, achieved a score of 98.54% on a publicly verifiable reconstruction of India's National Eligibility-cum-Entrance Test for Postgraduate medical courses (NEET PG) 2025 paper.

According to the company's evaluation, Zorgm Pro answered 152 of 154 questions correctly, with one incorrect answer and one unanswered question. The benchmark was conducted using a documented methodology that has been made publicly available for independent review.

The result demonstrates the potential of source-grounded clinical AI systems to perform at a high level on complex medical education benchmarks while maintaining transparency through evidence-backed responses.

Zorgm Pro is raising the bar for clinical AI

Unlike conventional AI systems that generate responses without source attribution, Zorgm Pro utilizes an end-to-end Retrieval Augmented Generation (RAG) architecture designed to reduce hallucinations. The platform retrieves information from a curated library of peer-reviewed journals, clinical guidelines, and regulator-approved drug labels, providing citations alongside responses.

The evaluation conducted by the LaennecAI Clinical Research Team found that Zorgm Pro performed strongly when assessed against the NEET PG benchmark and compared with other widely available AI systems evaluated under the same framework.

A notable example occurred on a question regarding paediatric HIV vertical-transmission prophylaxis, where Zorgm Pro cited India's national PPTCT guidance (NACO India), international NIH Perinatal Guidelines, and relevant FDA-approved references in support of its response.

"The results are impressive. The real win is achieving them with an AI that is less prone to hallucinations, one that looks up only curated, evidence-based sources, and is not a black-box system," said Dr. Akhil Das, Clinical Lead at LaennecAI.

"We built Zorgm Pro because doctors deserve AI they can trust," added Dr. Arathy Varghese, Co-founder and CTO of LaennecAI. "An answer is only as good as the evidence behind it. Ours comes with the source attached every time, so a doctor never has to take it on faith."

Elite medical education, democratized at zero cost

NEET PG is the primary gateway to specialist training for medical graduates in India, with up to 230,000 candidates competing each year. By offering a specialized clinical AI platform free for credentialed medical professionals, Zorgm Pro aims to support medical education, self-learning, and evidence-based reference.

In alignment with scientific transparency, the LaennecAI Clinical Research Team has published its technical methodology online. The corrected text-only question subset, prompt templates, and grading methodology have been made publicly accessible for independent review.

About LaennecAI

LaennecAI is a medical AI company with teams across London, Cardiff, Birmingham, and Kochi. The company builds clinically focused AI tools for doctors and patients under the Zorgm brand. Its products include Zorgm Pro, a free clinician-only medical education answer engine for continuing medical education, self-learning, and evidence-based reference, alongside patient-facing self-management technologies and an AI as Medical Device digital stethoscope for respiratory and cardiac monitoring.

LaennecAI's approach is grounded in curated medical sources, clinical governance, privacy, and regional relevance.

“This is a company press release that is not part of editorial content. No journalist of The Hindubusinessline was involved in the publication of this release.”

Published on June 24, 2026