Topology of Word Embeddings: Singularities Reflect Polysemy
Not on the map. Why?
Abstract (arXiv)
The manifold hypothesis suggests that word vectors live on a submanifold within their ambient vector space. We argue that we should, more accurately, expect them to live on a pinched manifold: a singular quotient of a manifold obtained by identifying some of its points. The identified, singular points correspond to polysemous words, i.e. words with multiple meanings. Our point of view suggests that monosemous and polysemous words can be distinguished based on the topology of their neighbourhoods. We present two kinds of empirical evidence to support this point of view: (1) We introduce a topological measure of polysemy based on persistent homology that correlates well with the actual number of meanings of a word. (2) We propose a simple, topologically motivated solution to the SemEval-2010 task on Word Sense Induction & Disambiguation that produces competitive results.
Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). First posted on arXiv 18 Nov 2020.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 1; 2022: 3; 2023: 0; 2024: 5; 2025: 5; 2026: 1.
15 papers.
Sort:
By area:
- 2021Let's Play Mono-Poly: BERT Can Reveal Words' Polysemy Level and Partitionability into Sensescited by 39
- 2022Acceptability Judgements via Examining the Topology of Attention Mapscited by 19
- 2025Token embeddings violate the manifold hypothesiscited by 14
- 2022Topological Singularity Detection at Multiple Scalescited by 11
- 2024The structure of the token space for large language modelscited by 10
- 2022Dialogue Term Extraction using Transfer Learning and Topological Data Analysiscited by 5
- 2025Probing the topology of the space of tokens with structured promptscited by 3
- 2024Local Topology Measures of Contextual Language Model Latent Spaces With Applications to Dialogue Term Extractioncited by 2
- 2024Topological Data Analysis Applications in Natural Language Processing: A Surveycited by 2
- 2025Exploring the Stratified Space Structure of an RL Game with the Volume Growth Transformcited by 1
- 2025Advancing mathematics research with generative AIcited by 1
- 2024LGDE: Local Graph-based Dictionary Expansioncited by 0
- 2026Text Corpora as Concept Fields: Black-Box Hallucination and Novelty Measurement6 May 2026
- 2025Advancing mathematics research with generative AI29 Sep 2025
- 2025Exploring the Stratified Space Structure of an RL Game with the Volume Growth Transform29 Jul 2025
- 2025Token embeddings violate the manifold hypothesis1 Apr 2025
- 2025The cell as a token: high-dimensional geometry in language models and cell embeddings26 Mar 2025
- 2025Probing the topology of the space of tokens with structured prompts19 Mar 2025
- 2024Topological Data Analysis Applications in Natural Language Processing: A Survey15 Nov 2024
- 2024The structure of the token space for large language models11 Oct 2024
- 2024Local Topology Measures of Contextual Language Model Latent Spaces With Applications to Dialogue Term Extraction7 Aug 2024
- 2024Topological quantification of ambiguity in semantic search12 Jun 2024
- 2024LGDE: Local Graph-based Dictionary Expansion13 May 2024
- 2022Topological Singularity Detection at Multiple Scales30 Sep 2022
- 2025Advancing mathematics research with generative AIcited by 1
- 2025Advancing mathematics research with generative AI29 Sep 2025
- 2021Let's Play Mono-Poly: BERT Can Reveal Words' Polysemy Level and Partitionability into Sensescited by 39
- 2022Acceptability Judgements via Examining the Topology of Attention Mapscited by 19
- 2025Token embeddings violate the manifold hypothesiscited by 14
- 2022Topological Singularity Detection at Multiple Scalescited by 11
- 2024The structure of the token space for large language modelscited by 10
- 2022Dialogue Term Extraction using Transfer Learning and Topological Data Analysiscited by 5
- 2025Probing the topology of the space of tokens with structured promptscited by 3
- 2024Local Topology Measures of Contextual Language Model Latent Spaces With Applications to Dialogue Term Extractioncited by 2
- 2024Topological Data Analysis Applications in Natural Language Processing: A Surveycited by 2
- 2025Exploring the Stratified Space Structure of an RL Game with the Volume Growth Transformcited by 1
- 2024LGDE: Local Graph-based Dictionary Expansioncited by 0
- 2024Topological quantification of ambiguity in semantic searchcited by 0
12 most cited of 14 All 14 (JSON)
- 2026Text Corpora as Concept Fields: Black-Box Hallucination and Novelty Measurement6 May 2026
- 2025Exploring the Stratified Space Structure of an RL Game with the Volume Growth Transform29 Jul 2025
- 2025Token embeddings violate the manifold hypothesis1 Apr 2025
- 2025The cell as a token: high-dimensional geometry in language models and cell embeddings26 Mar 2025
- 2025Probing the topology of the space of tokens with structured prompts19 Mar 2025
- 2024Topological Data Analysis Applications in Natural Language Processing: A Survey15 Nov 2024
- 2024The structure of the token space for large language models11 Oct 2024
- 2024Local Topology Measures of Contextual Language Model Latent Spaces With Applications to Dialogue Term Extraction7 Aug 2024
- 2024Topological quantification of ambiguity in semantic search12 Jun 2024
- 2024LGDE: Local Graph-based Dictionary Expansion13 May 2024
- 2022Topological Singularity Detection at Multiple Scales30 Sep 2022
- 2022Dialogue Term Extraction using Transfer Learning and Topological Data Analysis22 Aug 2022
12 newest of 14 All 14 (JSON)