Retrieval-Augmented Generation (RAG) is widely used for knowledge-intensive question answering (QA) but typically relies on large language models (LLMs) as generators, which raises cost and privacy concerns. We propose TripletRAG, a simple, training-free method that converts LLM-teacher behavior into triplets (synthetic question, retrieved passages, reference answer), which are then provided as exemplar context to a small language model (SLM) at inference. TripletRAG requires only offline synthetic QA generation and retrieval; no SLM finetuning is needed. Across SQuAD and Natural Questions (KILT) datasets, triplet guidance substantially increases judge-based accuracy, and improves groundedness on SQuAD for models in the 1B–3B parameter range, with negligible or inconsistent gains for larger models (7B+). Compared to a strong lightweight baseline (MiniRAG), TripletRAG gives markedly better groundedness and correctness while keeping the pipeline simple and practical.

We introduce SHARP, a compact, span-annotated dataset for hallucination detection in model-generated mathematical solutions. Unlike existing resources that rely on coarse step-level labels, SHARP provides fine-grained, span-level annotations generated through an automated pipeline. Our methodology synthesizes realistic reasoning traces with precise error markup, enabling the production of high-fidelity training data with minimal computational overhead. Despite being 10–30 tim smaller than existing open-source datasets, a 1.5B PRM fine-tuned on SHARP achieves state-of-the-art performance on ProcessBench.