mlsanityData to 7 Oct 2026

Take and Took, Gaggle and Goose, Book and Read: Evaluating the Utility of Vector Differences for Lexical Relation Learning

arXiv 1509.01692 ↗First posted 5 Sep 2015cs.CLcited by 29

Not on the map. Why?

Abstract (arXiv)

Recent work on word embeddings has shown that simple vector subtraction over pre-trained embeddings is surprisingly effective at capturing different lexical relations, despite lacking explicit supervision. Prior work has evaluated this intriguing result using a word analogy prediction formulation and hand-selected relations, but the generality of the finding over a broader range of lexical relation types and different learning settings has not been evaluated. In this paper, we carry out such an evaluation in two learning settings: (1) spectral clustering to induce word relations, and (2) supervised learning to classify vector differences into relation types. We find that word embeddings capture a surprising amount of information, and that, under suitable supervised training, vector subtraction generalises well to a broad range of relations, including over unseen lexical items.

Key facts. Cited by 29 papers in the map (data through 7 Oct 2026). First posted on arXiv 5 Sep 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 6; 2023: 4; 2024: 7; 2025: 5; 2026: 3.

29 papers.

Sort:

By area:

  1. 2023The Linear Representation Hypothesis and the Geometry of Large Language Modelscited by 679
  2. 2021BERT is to NLP what AlexNet is to CV: Can Pre-Trained Language Models Identify Analogies?cited by 56
  3. 2021Robust Retrieval Augmented Generation for Zero-shot Slot Fillingcited by 24
  4. 2024Does ChatGPT Have a Mind?cited by 13
  5. 2024Concept Alignmentcited by 9
  6. 2022Compression ensembles quantify aesthetic complexity and the evolution of visual artcited by 8
  7. 2022What do Toothbrushes do in the Kitchen? How Transformers Think our World is Structuredcited by 5
  8. 2023SEGA: Instructing Text-to-Image Models using Semantic Guidancecited by 4
  9. 2022Hierarchies over Vector Space: Orienting Word and Graph Embeddingscited by 3
  10. 2024A Multidimensional Framework for Evaluating Lexical Semantic Change with Social Science Applicationscited by 3
  11. 2024Leveraging Chemistry Foundation Models to Facilitate Structure Focused Retrieval Augmented Generation in Multi-Agent Workflows for Catalyst and Materials Designcited by 3
  12. 2021Paraphrases do not explain word analogiescited by 2
17 more
  1. 2021CO-NNECT: A Framework for Revealing Commonsense Knowledge Paths as Explicitations of Implicit Knowledge in Textscited by 2
  2. 2023RelBERT: Embedding Relations with Language Modelscited by 1
  3. 2023Semantic Guidance Tuning for Text-To-Image Diffusion Modelscited by 1
  4. 2024KnowledgePrompts: Exploring the Abilities of Large Language Models to Solve Proportional Analogies via Knowledge-Enhanced Promptingcited by 1
  5. 2025Linear Relational Decoding of Morphology in Language Modelscited by 1
  6. 2025Disentangling Latent Embeddings with Sparse Linear Concept Subspaces (SLiCS)cited by 1
  7. 2025Vocab Diet: Reshaping the Vocabulary of LLMs via Vector Arithmeticcited by 1
  8. 2022An Evaluation Dataset for Legal Word Embedding: A Case Study On Chinese Codexcited by 0
  9. 2022Assessing the Limits of the Distributional Hypothesis in Semantic Spaces: Trait-based Relational Knowledge and the Impact of Co-occurrencescited by 0
  10. 2022Beyond Model Interpretability: On the Faithfulness and Adversarial Robustness of Contrastive Textual Explanationscited by 0
  11. 2024Exploring Internal Numeracy in Language Models: A Case Study on ALBERTcited by 0
  12. 2024Optimal synthesis embeddingscited by 0
  13. 2025Gender Trouble in Language Models: An Empirical Audit Guided by Gender Performativity Theorycited by 0
  14. 2025AIR: Zero-shot Generative Model Adaptation with Iterative Refinementcited by 0
  15. 2026Open-SAT: LLM-Guided Query Embedding Refinement for Open-Vocabulary Object Retrieval in Satellite Imagerycited by 0
  16. 2026CAST: Closed-form Analytic Semantic Transfer for Zero-Shot Classifier Extensioncited by 0
  17. 2026Recovering Temporal and Geographic Signals from Language Model Embeddingscited by 0