mlsanityData to 7 Oct 2026

Rethinking Metrics for Lexical Semantic Change Detection

arXiv 2602.15716 ↗First posted 17 Feb 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Lexical semantic change detection (LSCD) increasingly relies on contextualised language model embeddings, yet most approaches still quantify change using a small set of semantic change metrics, primarily Average Pairwise Distance (APD) and cosine distance over word prototypes (PRT). We introduce Average Minimum Distance (AMD) and Symmetric Average Minimum Distance (SAMD), new measures that quantify semantic change via local correspondence between word usages across time periods. Across multiple languages, encoder models, and representation spaces, we show that AMD often provides more robust performance, particularly under dimensionality reduction and with non-specialised encoders, while SAMD excels with specialised encoders. We suggest that LSCD may benefit from considering alternative semantic change metrics beyond APD and PRT, with AMD offering a robust option for contextualised embedding-based analysis.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 42 references that are in the map. First posted on arXiv 17 Feb 2026.

Builds on

30 of 42 references.

References · 30

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  2. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,907
  3. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  4. 2024Multilingual E5 Text Embeddings: A Technical Reportcited by 357
  5. 2020Revisiting Pre-Trained Models for Chinese Natural Language Processingcited by 262
  6. 2016Diachronic Word Embeddings Reveal Statistical Laws of Semantic Changecited by 219
  7. 2020German's Next Language Modelcited by 144
  8. 2020SemEval-2020 Task 1: Unsupervised Lexical Semantic Change Detectioncited by 105
  9. 2020Analysing Lexical Semantic Change with Contextualised Word Representationscited by 92
  10. 2023Spanish Pre-trained BERT Model and Evaluation Datacited by 81
  11. 2016Cultural Shift or Linguistic Drift? Comparing Two Computational Measures of Semantic Changecited by 63
  12. 2014Temporal Analysis of Language through Neural Language Modelscited by 53
18 more
  1. 2014Improving zero-shot learning by mitigating the hubness problemcited by 53
  2. 2020Latin BERT: A Contextual Language Model for Classical Philologycited by 43
  3. 2025mmBERT: A Modern Multilingual Encoder with Annealed Language Learningcited by 39
  4. 2019Leveraging Contextual Embeddings for Detecting Diachronic Semantic Shiftcited by 37
  5. 2020Playing with Words at the National Library of Sweden -- Making a Swedish BERTcited by 35
  6. 2019Time-Out: Temporal Referencing for Robust Modeling of Lexical Semantic Changecited by 32
  7. 2020UiO-UvA at SemEval-2020 Task 1: Contextualised Embeddings for Lexical Semantic Change Detectioncited by 32
  8. 2019A Wind of Change: Detecting and Evaluating Lexical Semantic Change across Times and Domainscited by 31
  9. 2021Operationalizing a National Digital Library: The Case for a Norwegian Transformer Modelcited by 23
  10. 2024A Systematic Comparison of Contextualized Word Embeddings for Lexical Semantic Changecited by 21
  11. 2022Contextualized language models for semantic change detection: lessons learnedcited by 20
  12. 2021Lexical Semantic Change Discoverycited by 13
  13. 2022NorDiaChange: Diachronic Semantic Change Dataset for Norwegiancited by 7
  14. 2024Using Synchronic Definitions and Semantic Relations to Classify Semantic Change Typescited by 7
  15. 2024Definition generation for lexical semantic change detectioncited by 7
  16. 2025SenWiCh: Sense-Annotation of Low-Resource Languages for WiC using Hybrid Methodscited by 4
  17. 2025SCDTour: Embedding Axis Ordering and Merging for Interpretable Semantic Change Detectioncited by 4
  18. 2024The LSCD Benchmark: a Testbed for Diachronic Word Meaning Taskscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026When Lexical Change Misleads: Rethinking Dynamic Topic Model Evaluation with Traditional and LLM-Based Metrics
  2. 2024Semantic Change Characterization with LLMs using Rhetorics
  3. 2026MinCU: A Fine-Grained Benchmark for Grounded Minimal-Change Understanding in Image Pairs
  4. 2026C3-Bench: A Context-Aware Change Captioning Benchmark
  5. 2024Locally Measuring Cross-lingual Lexical Alignment: A Domain and Word Level Perspective
  6. 2025UniChange: Unifying Change Detection with Multimodal Large Language Model
  7. 2024VISLA Benchmark: Evaluating Embedding Sensitivity to Semantic and Lexical Alterations
  8. 2026Measuring Collective Semantic Change in Populations of Language Model Agents