@inproceedings{wang-etal-2026-unid3,
title = "{U}ni{D}$^{3}$: Open Knowledge Graphs and {QA} Benchmarks for Drug{--}Disease Discovery and Reasoning",
author = "Wang, Qing and
Liu, Tianshi and
Zhou, Minghao and
Liang, Jialu and
Guo, Sen and
Su, Jing and
Song, Qianqian",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.588/",
pages = "12399--12421",
ISBN = "979-8-89176-462-0",
abstract = "Drug{--}disease knowledge underpins drug discovery and precision medicine, yet most of it is locked in unstructured PubMed articles and only partially captured by curated databases. We release UniD{\textthreesuperior}-KG, a suite of six knowledge graphs spanning three tasks (Drug{--}Disease Matching, DDM; Drug Effectiveness Assessment, DEA; and Drug{--}Target Analysis, DTA), and UniD{\textthreesuperior}-data, three open QA datasets (28,915 DDM, 15,042 DEA and {\ensuremath{>}}4,000 DTA pairs), distilled from 157,849 PubMed articles with Llama-3.3-70B by the UniD{\textthreesuperior} KG-RAG pipeline, which couples task-specific prompting with a dual-stage entity extraction that re-extracts over the consolidated graph to recover cross-chunk and cross-document evidence. External validation on CDR, DrugReviews, P3ps and MedicationQA and a benchmark across seven LLM baselines and three retrieval variants show that UniD{\textthreesuperior}-data is faithful to curated sources and that retrieval over UniD{\textthreesuperior}-KG consistently improves biomedical QA; a clinician verification of 469 FDA-approved, US-marketed drugs further confirms strong agreement with the UniD{\textthreesuperior}-data DDM split (accuracy 0.90, F1 0.90, AUROC 0.90). All artefacts{---}UniD{\textthreesuperior}-KG, UniD{\textthreesuperior}-data, benchmark scripts and the UniD{\textthreesuperior}-QA explainable chatbot{---}are released as shared infrastructure for biomedical NLP and drug repurposing."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-unid3">
<titleInfo>
<title>UniD³: Open Knowledge Graphs and QA Benchmarks for Drug–Disease Discovery and Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Qing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianshi</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghao</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jialu</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sen</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jing</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qianqian</namePart>
<namePart type="family">Song</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>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Drug–disease knowledge underpins drug discovery and precision medicine, yet most of it is locked in unstructured PubMed articles and only partially captured by curated databases. We release UniD³-KG, a suite of six knowledge graphs spanning three tasks (Drug–Disease Matching, DDM; Drug Effectiveness Assessment, DEA; and Drug–Target Analysis, DTA), and UniD³-data, three open QA datasets (28,915 DDM, 15,042 DEA and \ensuremath>4,000 DTA pairs), distilled from 157,849 PubMed articles with Llama-3.3-70B by the UniD³ KG-RAG pipeline, which couples task-specific prompting with a dual-stage entity extraction that re-extracts over the consolidated graph to recover cross-chunk and cross-document evidence. External validation on CDR, DrugReviews, P3ps and MedicationQA and a benchmark across seven LLM baselines and three retrieval variants show that UniD³-data is faithful to curated sources and that retrieval over UniD³-KG consistently improves biomedical QA; a clinician verification of 469 FDA-approved, US-marketed drugs further confirms strong agreement with the UniD³-data DDM split (accuracy 0.90, F1 0.90, AUROC 0.90). All artefacts—UniD³-KG, UniD³-data, benchmark scripts and the UniD³-QA explainable chatbot—are released as shared infrastructure for biomedical NLP and drug repurposing.</abstract>
<identifier type="citekey">wang-etal-2026-unid3</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.588/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>12399</start>
<end>12421</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T UniD³: Open Knowledge Graphs and QA Benchmarks for Drug–Disease Discovery and Reasoning
%A Wang, Qing
%A Liu, Tianshi
%A Zhou, Minghao
%A Liang, Jialu
%A Guo, Sen
%A Su, Jing
%A Song, Qianqian
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F wang-etal-2026-unid3
%X Drug–disease knowledge underpins drug discovery and precision medicine, yet most of it is locked in unstructured PubMed articles and only partially captured by curated databases. We release UniD³-KG, a suite of six knowledge graphs spanning three tasks (Drug–Disease Matching, DDM; Drug Effectiveness Assessment, DEA; and Drug–Target Analysis, DTA), and UniD³-data, three open QA datasets (28,915 DDM, 15,042 DEA and \ensuremath>4,000 DTA pairs), distilled from 157,849 PubMed articles with Llama-3.3-70B by the UniD³ KG-RAG pipeline, which couples task-specific prompting with a dual-stage entity extraction that re-extracts over the consolidated graph to recover cross-chunk and cross-document evidence. External validation on CDR, DrugReviews, P3ps and MedicationQA and a benchmark across seven LLM baselines and three retrieval variants show that UniD³-data is faithful to curated sources and that retrieval over UniD³-KG consistently improves biomedical QA; a clinician verification of 469 FDA-approved, US-marketed drugs further confirms strong agreement with the UniD³-data DDM split (accuracy 0.90, F1 0.90, AUROC 0.90). All artefacts—UniD³-KG, UniD³-data, benchmark scripts and the UniD³-QA explainable chatbot—are released as shared infrastructure for biomedical NLP and drug repurposing.
%U https://aclanthology.org/2026.findings-emnlp.588/
%P 12399-12421
Markdown (Informal)
[UniD3: Open Knowledge Graphs and QA Benchmarks for Drug–Disease Discovery and Reasoning](https://aclanthology.org/2026.findings-emnlp.588/) (Wang et al., Findings 2026)
ACL