A unified multi-domain framework for hallucination detection and reliability evaluation in large language models

Summary

Researchers introduce MDH-Bench, a benchmark of 400 adversarial yet naturalistic prompts across eight reliability categories, and a Unified Multi-Domain Reliability Evaluation Framework that combines accuracy, hallucination rate, contradiction rate and safety indicators into a composite R-Score. Across four LLMs (GPT-5.2, Claude Sonnet 4, Gemini 2.5 Pro, DeepSeek V3.1), GPT-5.2 scored highest (80.10% accuracy, R-Score 0.8715), while DeepSeek V3.1 scored lowest (63.60% accuracy, 37.55% hallucination rate, R-Score 0.7074).

Related items

• InfoThe power of AI is worrying senior Labour figures – as is the fear of ‘falling behind the public’Same vendor · The Guardian Technology
• InfoQuoting Addy Osmani, Morgan Westlee Lunt, and Qing WangSame vendor · Simon Willison's Weblog
• HighCVE-2026-96620: Claude Agent SDK argument injection via resume option enabling command executionSame vendor · NVD/CVE Database
• MediumFake Claude, ChatGPT Attacks Surge, Putting Developers at RiskSame vendor · Dark Reading
• InfoHow Itaú is building generative AI for its entire customer baseSame vendor · OpenAI Blog