@inproceedings{abdalla-etal-2026-obliq,
title = "{OBLIQ}-{IR}: Training a Dense Retriever for Oblique Queries",
author = "Abdalla, Mahmoud and
Abdallah, Abdelrahman and
Sedek, Shaimaa and
Jatowt, Adam",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.494/",
pages = "11348--11369",
ISBN = "979-8-89176-459-0",
abstract = "Oblique retrieval, recently introduced by OBLIQ-Bench, asks a retriever to surface documents whose relevance is determined by a latent attribute (an implicit stance, an analogous reasoning technique, an authorial fingerprint, or a vague tip-of-the-tongue recollection) that has little or no surface expression in the document. State-of-the-art dense encoders and agentic search pipelines built around frontier language models leave a large first-stage bottleneck on these tasks, while the same language models reliably verify relevance when shown candidates. We address this with OBLIQ-IR, a single-vector dense retriever whose training mixture combines per-mechanism synthetic queries with a new form of cross-model supervision: kNN-graph distillation from a frozen authorship encoder, which transfers a style-versus-topic inductive bias into the student. A 3B retriever fine-tuned reaches 0.211 NDCG@10 on Writing-Style, 0.171 on Math, 0.177 on Twitter, and 0.281 on Congress, improving over the GPT-5.2 Multi-Hop Agent by 0.010 to 0.098 NDCG@10 and over Gemini-2-Embedding by 0.027 to 0.222 NDCG@10 on every reported task."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="abdalla-etal-2026-obliq">
<titleInfo>
<title>OBLIQ-IR: Training a Dense Retriever for Oblique Queries</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mahmoud</namePart>
<namePart type="family">Abdalla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdelrahman</namePart>
<namePart type="family">Abdallah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shaimaa</namePart>
<namePart type="family">Sedek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adam</namePart>
<namePart type="family">Jatowt</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 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Oblique retrieval, recently introduced by OBLIQ-Bench, asks a retriever to surface documents whose relevance is determined by a latent attribute (an implicit stance, an analogous reasoning technique, an authorial fingerprint, or a vague tip-of-the-tongue recollection) that has little or no surface expression in the document. State-of-the-art dense encoders and agentic search pipelines built around frontier language models leave a large first-stage bottleneck on these tasks, while the same language models reliably verify relevance when shown candidates. We address this with OBLIQ-IR, a single-vector dense retriever whose training mixture combines per-mechanism synthetic queries with a new form of cross-model supervision: kNN-graph distillation from a frozen authorship encoder, which transfers a style-versus-topic inductive bias into the student. A 3B retriever fine-tuned reaches 0.211 NDCG@10 on Writing-Style, 0.171 on Math, 0.177 on Twitter, and 0.281 on Congress, improving over the GPT-5.2 Multi-Hop Agent by 0.010 to 0.098 NDCG@10 and over Gemini-2-Embedding by 0.027 to 0.222 NDCG@10 on every reported task.</abstract>
<identifier type="citekey">abdalla-etal-2026-obliq</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.494/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>11348</start>
<end>11369</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T OBLIQ-IR: Training a Dense Retriever for Oblique Queries
%A Abdalla, Mahmoud
%A Abdallah, Abdelrahman
%A Sedek, Shaimaa
%A Jatowt, Adam
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F abdalla-etal-2026-obliq
%X Oblique retrieval, recently introduced by OBLIQ-Bench, asks a retriever to surface documents whose relevance is determined by a latent attribute (an implicit stance, an analogous reasoning technique, an authorial fingerprint, or a vague tip-of-the-tongue recollection) that has little or no surface expression in the document. State-of-the-art dense encoders and agentic search pipelines built around frontier language models leave a large first-stage bottleneck on these tasks, while the same language models reliably verify relevance when shown candidates. We address this with OBLIQ-IR, a single-vector dense retriever whose training mixture combines per-mechanism synthetic queries with a new form of cross-model supervision: kNN-graph distillation from a frozen authorship encoder, which transfers a style-versus-topic inductive bias into the student. A 3B retriever fine-tuned reaches 0.211 NDCG@10 on Writing-Style, 0.171 on Math, 0.177 on Twitter, and 0.281 on Congress, improving over the GPT-5.2 Multi-Hop Agent by 0.010 to 0.098 NDCG@10 and over Gemini-2-Embedding by 0.027 to 0.222 NDCG@10 on every reported task.
%U https://aclanthology.org/2026.emnlp-main.494/
%P 11348-11369
Markdown (Informal)
[OBLIQ-IR: Training a Dense Retriever for Oblique Queries](https://aclanthology.org/2026.emnlp-main.494/) (Abdalla et al., EMNLP 2026)
ACL
- Mahmoud Abdalla, Abdelrahman Abdallah, Shaimaa Sedek, and Adam Jatowt. 2026. OBLIQ-IR: Training a Dense Retriever for Oblique Queries. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 11348–11369, Budapest, Hungary. Association for Computational Linguistics.