analiza aiqbench
This arXiv preprint proposes a student-teacher, human-in-the-loop LLM pipeline for checking pharmaceutical content against regulatory, brand, and technical rules. It reports 83.0% F1 and 97.5% recall on AIReg Bench, but the comparison is against a single model (Gemini 2.5 Pro), and the authors concede weak recall on grammatical (25.0%) and punctuation (41.7%) errors. Readers should treat the results as promising for a narrow compliance task, with the benchmark design and the value of the dual-model setup still needing independent replication.
Are Two LLMs Better Than One? A Student-Teacher Dual-Head LLMs Architecture for Pharmaceutical Content Optimization
arXiv:2602.11957v2 Announce Type: replace Abstract: Large Language Models (LLMs) are increasingly used for content creation in regulated domains such as pharmaceuticals, where content must be medical accurate and legally compliant. Traditional manual quality control (QC) is slow, error-prone, and create publication bottlenecks. To address this, we introduce a modular, LLM/VLMs-driven QC architecture, LRBTC (Language, Regulatory, Brand, Technical, Content Structure), implemented through a Student Teacher HITL architecture and Waterfall rule filtering logic for scalable, verifiable content validation. This architecture is designed to provide verifiable, traceable, and scalable content optimization. Our methods achieve 83.0% F1 and 97.5% recall on AIReg Bench, reducing missed violations five fold compared with Gemini 2.5 Pro. On CSpelling, it improves mean accuracy by +26.7 %. Our error analysis further reveals that while current models are strong at detecting misspellings (92.5% recall), they fail to identify complex grammatical (25.0% recall) and punctuation (41.7% recall) errors, highlighting a key area for future work. This work provides a practical, plug-and-play solution for reliable, transparent quality control of content in high-stakes, compliance-critical industries. We also provide demo access.
Title:Are Two LLMs Better Than One? A Student-Teacher Dual-Head LLMs Architecture for Pharmaceutical Content Optimization
Abstract:Large Language Models (LLMs) are increasingly used for content creation in regulated domains such as pharmaceuticals, where content must be medical accurate and legally compliant. Traditional manual quality control (QC) is slow, error-prone, and create publication bottlenecks. To address this, we introduce a modular, LLM/VLMs-driven QC architecture, LRBTC (Language, Regulatory, Brand, Technical, Content Structure), implemented through a Student Teacher HITL architecture and Waterfall rule filtering logic for scalable, verifiable content validation. This architecture is designed to provide verifiable, traceable, and scalable content optimization. Our methods achieve 83.0% F1 and 97.5% recall on AIReg Bench, reducing missed violations five fold compared with Gemini 2.5 Pro. On CSpelling, it improves mean accuracy by +26.7 %. Our error analysis further reveals that while current models are strong at detecting misspellings (92.5% recall), they fail to identify complex grammatical (25.0% recall) and punctuation (41.7% recall) errors, highlighting a key area for future work. This work provides a practical, plug-and-play solution for reliable, transparent quality control of content in high-stakes, compliance-critical industries. We also provide demo access.
Machine Learning (cs.LG)
arXiv:2602.11957 [cs.LG]
(or arXiv:2602.11957v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2602.11957 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 279 cuvinte