@inproceedings{kim-joe-2026-peeling,
title = "Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings",
author = "Kim, Eunchan and
Joe, Inwhee",
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.1305/",
pages = "29717--29729",
ISBN = "979-8-89176-459-0",
abstract = "Taxonomy construction is essential for knowledge-intensive NLP yet remains costly: clustering-based methods (i) need careful per-corpus tuning and yield unstable hierarchies, while LLM-based pipelines (ii) depend on expensive task-specific training that does not transfer across domains. We report an empirical finding that bears on both: in Matryoshka Representation Learning (MRL) embeddings, where each leading prefix is itself a usable, coarser vector, prefix length aligns with the coarse-to-fine granularity of gold taxonomies; shuffled- and reverse-dimension controls isolate the alignment to the learned early prefix dimensions. The pattern holds across three architecturally distinct MRL models (full statistics in {\textsection}3). To test whether this regularity is usable, we introduce PEEL (Peeling Embeddings to Extract Latent ontologies): a minimal probe over these frozen embeddings, trained once with an order-embedding objective that represents hierarchy as vector containment; a minimum spanning arborescence then decodes taxonomic DAGs. Using only a concept vocabulary (no documents), PEEL recovers taxonomic DAGs with high structural fidelity on two OLLM benchmarks, transferring to a new domain with no target-edge supervision. The end-to-end OLLM pipeline{---}which jointly discovers concepts and induces structure from document corpora{---}is reported as a cross-task reference ({\textsection}5.3). Pre-trained retrieval embeddings encode a decodable granularity gradient; a minimal probe renders it as explicit parent{--}child (is-a) structure."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kim-joe-2026-peeling">
<titleInfo>
<title>Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings</title>
</titleInfo>
<name type="personal">
<namePart type="given">Eunchan</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Inwhee</namePart>
<namePart type="family">Joe</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>Taxonomy construction is essential for knowledge-intensive NLP yet remains costly: clustering-based methods (i) need careful per-corpus tuning and yield unstable hierarchies, while LLM-based pipelines (ii) depend on expensive task-specific training that does not transfer across domains. We report an empirical finding that bears on both: in Matryoshka Representation Learning (MRL) embeddings, where each leading prefix is itself a usable, coarser vector, prefix length aligns with the coarse-to-fine granularity of gold taxonomies; shuffled- and reverse-dimension controls isolate the alignment to the learned early prefix dimensions. The pattern holds across three architecturally distinct MRL models (full statistics in §3). To test whether this regularity is usable, we introduce PEEL (Peeling Embeddings to Extract Latent ontologies): a minimal probe over these frozen embeddings, trained once with an order-embedding objective that represents hierarchy as vector containment; a minimum spanning arborescence then decodes taxonomic DAGs. Using only a concept vocabulary (no documents), PEEL recovers taxonomic DAGs with high structural fidelity on two OLLM benchmarks, transferring to a new domain with no target-edge supervision. The end-to-end OLLM pipeline—which jointly discovers concepts and induces structure from document corpora—is reported as a cross-task reference (§5.3). Pre-trained retrieval embeddings encode a decodable granularity gradient; a minimal probe renders it as explicit parent–child (is-a) structure.</abstract>
<identifier type="citekey">kim-joe-2026-peeling</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1305/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>29717</start>
<end>29729</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings
%A Kim, Eunchan
%A Joe, Inwhee
%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 kim-joe-2026-peeling
%X Taxonomy construction is essential for knowledge-intensive NLP yet remains costly: clustering-based methods (i) need careful per-corpus tuning and yield unstable hierarchies, while LLM-based pipelines (ii) depend on expensive task-specific training that does not transfer across domains. We report an empirical finding that bears on both: in Matryoshka Representation Learning (MRL) embeddings, where each leading prefix is itself a usable, coarser vector, prefix length aligns with the coarse-to-fine granularity of gold taxonomies; shuffled- and reverse-dimension controls isolate the alignment to the learned early prefix dimensions. The pattern holds across three architecturally distinct MRL models (full statistics in §3). To test whether this regularity is usable, we introduce PEEL (Peeling Embeddings to Extract Latent ontologies): a minimal probe over these frozen embeddings, trained once with an order-embedding objective that represents hierarchy as vector containment; a minimum spanning arborescence then decodes taxonomic DAGs. Using only a concept vocabulary (no documents), PEEL recovers taxonomic DAGs with high structural fidelity on two OLLM benchmarks, transferring to a new domain with no target-edge supervision. The end-to-end OLLM pipeline—which jointly discovers concepts and induces structure from document corpora—is reported as a cross-task reference (§5.3). Pre-trained retrieval embeddings encode a decodable granularity gradient; a minimal probe renders it as explicit parent–child (is-a) structure.
%U https://aclanthology.org/2026.emnlp-main.1305/
%P 29717-29729
Markdown (Informal)
[Peeling the Matryoshka: Extracting Ontological Structure from Retrieval Embeddings](https://aclanthology.org/2026.emnlp-main.1305/) (Kim & Joe, EMNLP 2026)
ACL