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.

This paper deals with the task of annotating open-domain conversations with speech functions. We propose a semi-automated method for annotating dialogs following the topic-oriented, multi-layered taxonomy of speech functions with the use of hierarchical guidelines using Large Language Models. These guidelines comprise simple questions about the topic and speaker change, sentence types, pragmatic aspects of the utterance, and examples that aid untrained annotators in understanding the taxonomy. We compare the results of dialog annotation performed by experts, crowdsourcing workers, and ChatGPT. To improve the performance of ChatGPT, several experiments utilising different prompt engineering techniques were conducted. We demonstrate that in some cases large language models can achieve human-like performance following a multi-step tree-like annotation pipeline on complex discourse annotation, which is usually challenging and costly in terms of time and money when performed by humans.