@inproceedings{sullivan-pao-etal-2026-distill,
title = "Distill-{RE}: Few-Shot Relation Extraction via Teacher {LLM} Reasoning Distillation",
author = "Sullivan-Pao, Danielle and
Alemany, Sheila and
Simek, Olga",
editor = "Kalo, Jan-Christoph and
Nakashole, Ndapa and
Razniewski, Simon and
Suchanek, Fabian M. and
Vlachos, Andreas and
McCallum, Andrew",
booktitle = "Proceedings of the 11th Workshop on Automated Knowledge Base Construction ({AKBC} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.akbc-1.25/",
pages = "359--378",
ISBN = "979-8-89176-464-4",
abstract = "Despite their generalization capabilities, large language models (LLMs) have counterintuitively underperformed smaller pre-trained language models (PLMs) on few-shot relation extraction (RE), the task of identifying semantic relationships between entities in text. This gap is primarily due to a systematic over-prediction bias in LLMs. We propose Distill-RE, a two-stage knowledge distillation framework in which a teacher LLM generates chain-of-thought reasoning over existing training examples to supervise two student models: a classification model that predicts relations and a confirmation model that filters false positives. Across student scales from 3B to 70B, Distill-RE establishes a new state-of-the-art among RE models, with its top-performing configuration outperforming the strongest baseline by over 31.5 micro-F1 points averaged across TACRED, TACREV, and Re-TACRED at 8-, 16-, and 32-shot settings. Notably, without synthesizing new relation instances, it outperforms most data augmentation baselines and comes within 1.8 micro-F1 points on average of the strongest."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sullivan-pao-etal-2026-distill">
<titleInfo>
<title>Distill-RE: Few-Shot Relation Extraction via Teacher LLM Reasoning Distillation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Danielle</namePart>
<namePart type="family">Sullivan-Pao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sheila</namePart>
<namePart type="family">Alemany</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Olga</namePart>
<namePart type="family">Simek</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 11th Workshop on Automated Knowledge Base Construction (AKBC 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jan-Christoph</namePart>
<namePart type="family">Kalo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ndapa</namePart>
<namePart type="family">Nakashole</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Razniewski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabian</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Suchanek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrew</namePart>
<namePart type="family">McCallum</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">979-8-89176-464-4</identifier>
</relatedItem>
<abstract>Despite their generalization capabilities, large language models (LLMs) have counterintuitively underperformed smaller pre-trained language models (PLMs) on few-shot relation extraction (RE), the task of identifying semantic relationships between entities in text. This gap is primarily due to a systematic over-prediction bias in LLMs. We propose Distill-RE, a two-stage knowledge distillation framework in which a teacher LLM generates chain-of-thought reasoning over existing training examples to supervise two student models: a classification model that predicts relations and a confirmation model that filters false positives. Across student scales from 3B to 70B, Distill-RE establishes a new state-of-the-art among RE models, with its top-performing configuration outperforming the strongest baseline by over 31.5 micro-F1 points averaged across TACRED, TACREV, and Re-TACRED at 8-, 16-, and 32-shot settings. Notably, without synthesizing new relation instances, it outperforms most data augmentation baselines and comes within 1.8 micro-F1 points on average of the strongest.</abstract>
<identifier type="citekey">sullivan-pao-etal-2026-distill</identifier>
<location>
<url>https://aclanthology.org/2026.akbc-1.25/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>359</start>
<end>378</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Distill-RE: Few-Shot Relation Extraction via Teacher LLM Reasoning Distillation
%A Sullivan-Pao, Danielle
%A Alemany, Sheila
%A Simek, Olga
%Y Kalo, Jan-Christoph
%Y Nakashole, Ndapa
%Y Razniewski, Simon
%Y Suchanek, Fabian M.
%Y Vlachos, Andreas
%Y McCallum, Andrew
%S Proceedings of the 11th Workshop on Automated Knowledge Base Construction (AKBC 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-464-4
%F sullivan-pao-etal-2026-distill
%X Despite their generalization capabilities, large language models (LLMs) have counterintuitively underperformed smaller pre-trained language models (PLMs) on few-shot relation extraction (RE), the task of identifying semantic relationships between entities in text. This gap is primarily due to a systematic over-prediction bias in LLMs. We propose Distill-RE, a two-stage knowledge distillation framework in which a teacher LLM generates chain-of-thought reasoning over existing training examples to supervise two student models: a classification model that predicts relations and a confirmation model that filters false positives. Across student scales from 3B to 70B, Distill-RE establishes a new state-of-the-art among RE models, with its top-performing configuration outperforming the strongest baseline by over 31.5 micro-F1 points averaged across TACRED, TACREV, and Re-TACRED at 8-, 16-, and 32-shot settings. Notably, without synthesizing new relation instances, it outperforms most data augmentation baselines and comes within 1.8 micro-F1 points on average of the strongest.
%U https://aclanthology.org/2026.akbc-1.25/
%P 359-378
Markdown (Informal)
[Distill-RE: Few-Shot Relation Extraction via Teacher LLM Reasoning Distillation](https://aclanthology.org/2026.akbc-1.25/) (Sullivan-Pao et al., AKBC 2026)
ACL