MAM-AI: An On-Device Medical Retrieval-Augmented Generation System for Nurses and Midwives in Zanzibar
On the map: Retrieval-augmented generation → Medical and biomedical RAG
Abstract (arXiv)
Maternal and newborn mortality remain among the highest in sub-Saharan Africa, where midwifery care is often delivered by nurses who lack midwifery training to international standards, and consulting authoritative guidance at the point of care is hard: the guidelines are long and connectivity is intermittent. We present MAM-AI, a medical question-answering assistant for nurse-midwives in Zanzibar that runs entirely on a commodity Android device: a question is embedded (EmbeddingGemma, 300M) and matched against a curated corpus of 87 guideline documents (63,650 passages), then answered with citations by a 4B int4 generator (Gemma 4 E4B), fully offline, with no query leaving the device. We evaluate the exact deployed configuration with a layered methodology -- retriever, generator under oracle context, end-to-end, and latency -- scored by LLM judges validated against physician rubrics. The evaluation relocates the hard problem. On-device retrieval is essentially solved: the 300M embedder ranks third of seven retrievers and rivals cloud systems, so the passages the system needs are usually found. The small generator is what remains in doubt: adding retrieved context does not improve its answers, and at 4B it cannot be both helpful and safe at once -- of two same-size candidates, the more helpful one commits genuine dangerous errors, so we deploy the other, which is about twice as faithful to its sources (as faithful as a frontier model), and recover its helpfulness with a redesigned prompt that cuts deflection from 33% to 3%. Corpus quality is decisive for the same reason: where the corpus holds the right passage the answer is specific and actionable, and where it does not it goes vague. MAM-AI is a thoroughly evaluated, open-source research prototype, not a fielded product; the system, knowledge base, benchmarks, and evaluation harness are released.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Medical and biomedical RAG, one of 15 directions in Retrieval-augmented generation. Builds on 10 of its 25 references that are in the map. First posted on arXiv 28 Jun 2026.
Builds on
10 of 25 references.
Closest ancestors · 1
Same direction, newest first
Other references · 9
- 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
- 2022LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scalecited by 906
- 2025HealthBench: Evaluating Large Language Models Towards Improved Human Healthcited by 429
- 2024MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documentscited by 257
- 2023MedCPT: Contrastive Pre-trained Transformers with Large-scale PubMed Search Logs for Zero-shot Biomedical Information Retrievalcited by 185
- 2025EmbeddingGemma: Powerful and Lightweight Text Representationscited by 139
- 2024Lynx: An Open Source Hallucination Evaluation Modelcited by 57
- 2024AfriMed-QA: A Pan-African, Multi-Specialty, Medical Question-Answering Benchmark Datasetcited by 26
- 2024MedAide: Leveraging Large Language Models for On-Premise Medical Assistance on Edge Devicescited by 9
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026A corpus-specific clinical RAG system matches or outperforms newer frontier LLMs on HealthBench
- 2024The Geometry of Queries: Query-Based Innovations in Retrieval-Augmented Generation for Healthcare QA
- 2024Comprehensive and Practical Evaluation of Retrieval-Augmented Generation Systems for Medical Question Answering
- 2025Retrieval-Augmented Clinical Benchmarking for Contextual Model Testing in Kenyan Primary Care: A Methodology Paper
- 2026mamabench and mamaretrieval: Benchmarks for Evaluating Medical Retrieval-Augmented Generation in Maternal, Neonatal, and Reproductive Health
- 2025Efficient and Reproducible Biomedical Question Answering using Retrieval Augmented Generation
- 2023MKRAG: Medical Knowledge Retrieval Augmented Generation for Medical Question Answering
- 2025IMB: An Italian Medical Benchmark for Question Answering