{"slug": "safellm-extraction-as-a-hallucination-resistant-alternative-to-rewriting-in", "title": "SafeLLM: Extraction as a Hallucination-Resistant Alternative to Rewriting in Safety-Critical Settings", "summary": "Researchers at arXiv have proposed SafeLLM, an extraction-based method for retrieval-augmented generation that eliminates hallucinations by directly selecting relevant sentences from source documents rather than rewriting content. In tests on NHS and NICE clinical guidelines, line-number-based source selection achieved up to 95% term recall while maintaining close alignment with original text, outperforming both direct copying and safety-focused rewriting strategies. The findings suggest that extraction-based approaches offer a more reliable alternative for deploying LLMs in safety-critical settings such as healthcare and compliance, where factual accuracy is paramount.", "body_md": "arXiv:2606.12897v1 Announce Type: new\nAbstract: Large language models (LLMs) are increasingly used to access organisational documentation, including standard operating procedures (SOPs), HR policies and institutional guidelines. However, retrieval-augmented generation (RAG) systems that rely on free-form rewriting can introduce hallucinations and unstable trade-offs between completeness and conciseness, particularly in safety- and compliance-critical settings. Objectives: To evaluate extraction as a hallucination-resistant alternative to rewriting-based RAG and compare strategies that balance precision, recall and safety across document types and model scales. Methods: We compare multiple prompting strategies, including line-number-based source selection, extraction of relevant guideline sentences with explicit safety annotations, and a multi-stage pipeline that refines draft answers using supporting evidence from source guidelines. Experiments are conducted on documents of varying length and structure, including local NHS acute care and oncology guidelines and UK-wide NICE guidelines, using both frontier-scale and locally deployable models. Performance is assessed using automatic metrics and human expert evaluation of relevance and completeness. Results: Line-number selection achieves the strongest results, outperforming direct copying and safety-focused strategies across both large and small models while maintaining high term recall (up to 95%) and close alignment with source text. Safety-oriented approaches improve precision but introduce systematic omissions, while multi-stage filtering further amplifies this trade-off. Performance varies with document structure: line-based extraction excels in protocol-like content, whereas alternative strategies perform better on more verbose documents (up to 97% term recall).", "url": "https://wpnews.pro/news/safellm-extraction-as-a-hallucination-resistant-alternative-to-rewriting-in", "canonical_source": "https://arxiv.org/abs/2606.12897", "published_at": "2026-06-12 04:00:00+00:00", "updated_at": "2026-06-12 04:56:39.807377+00:00", "lang": "en", "topics": ["large-language-models", "ai-safety", "natural-language-processing", "ai-research"], "entities": ["NHS", "NICE"], "alternates": {"html": "https://wpnews.pro/news/safellm-extraction-as-a-hallucination-resistant-alternative-to-rewriting-in", "markdown": "https://wpnews.pro/news/safellm-extraction-as-a-hallucination-resistant-alternative-to-rewriting-in.md", "text": "https://wpnews.pro/news/safellm-extraction-as-a-hallucination-resistant-alternative-to-rewriting-in.txt", "jsonld": "https://wpnews.pro/news/safellm-extraction-as-a-hallucination-resistant-alternative-to-rewriting-in.jsonld"}}