@inproceedings{tambovtsev-etal-2026-enhancing,
title = "Enhancing Small Language Models in {RAG} via Triplet-Guided In-Context Distillation",
author = "Tambovtsev, Ilia and
Molchanova, Maria and
Mikhailova, Anna and
Korotkov, Alexey and
Sgibnev, Ivan and
Vizilter, Iurii and
Savchenko, Andrey",
editor = "Wang, Yimu and
Choi, Yee Man and
Wu, Di and
Ouyang, Siqi and
Azadani, Mozhgan Nasr and
Fung, Yi R. (May)",
booktitle = "Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently ({G}round{LM} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.groundlm-1.72/",
pages = "1001--1015",
ISBN = "979-8-89176-478-1",
abstract = "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."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tambovtsev-etal-2026-enhancing">
<titleInfo>
<title>Enhancing Small Language Models in RAG via Triplet-Guided In-Context Distillation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ilia</namePart>
<namePart type="family">Tambovtsev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Molchanova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Mikhailova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexey</namePart>
<namePart type="family">Korotkov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="family">Sgibnev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iurii</namePart>
<namePart type="family">Vizilter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrey</namePart>
<namePart type="family">Savchenko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yimu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yee</namePart>
<namePart type="given">Man</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siqi</namePart>
<namePart type="family">Ouyang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mozhgan</namePart>
<namePart type="given">Nasr</namePart>
<namePart type="family">Azadani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="given">R</namePart>
<namePart type="given">(May)</namePart>
<namePart type="family">Fung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-478-1</identifier>
</relatedItem>
<abstract>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.</abstract>
<identifier type="citekey">tambovtsev-etal-2026-enhancing</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.72/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1001</start>
<end>1015</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Enhancing Small Language Models in RAG via Triplet-Guided In-Context Distillation
%A Tambovtsev, Ilia
%A Molchanova, Maria
%A Mikhailova, Anna
%A Korotkov, Alexey
%A Sgibnev, Ivan
%A Vizilter, Iurii
%A Savchenko, Andrey
%Y Wang, Yimu
%Y Choi, Yee Man
%Y Wu, Di
%Y Ouyang, Siqi
%Y Azadani, Mozhgan Nasr
%Y Fung, Yi R. (May)
%S Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-478-1
%F tambovtsev-etal-2026-enhancing
%X 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.
%U https://aclanthology.org/2026.groundlm-1.72/
%P 1001-1015
Markdown (Informal)
[Enhancing Small Language Models in RAG via Triplet-Guided In-Context Distillation](https://aclanthology.org/2026.groundlm-1.72/) (Tambovtsev et al., GroundLM 2026)
ACL
- Ilia Tambovtsev, Maria Molchanova, Anna Mikhailova, Alexey Korotkov, Ivan Sgibnev, Iurii Vizilter, and Andrey Savchenko. 2026. Enhancing Small Language Models in RAG via Triplet-Guided In-Context Distillation. In Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026), pages 1001–1015, Budapest, Hungary. Association for Computational Linguistics.