Rethinking Metrics for Lexical Semantic Change Detection

Research Summary

Contribution Summary

• 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
• 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

Related Papers

• Pushing the Frontier of Black-Box LVLM Attacks via Fine-Grained Detail Targeting Embedding Neighbor
• Axis Decomposition for ODRL: Resolving Dimensional Ambiguity in Policy Constraints through Interval Semantics Embedding Neighbor
• Diagnosing LLM Reranker Behavior Under Fixed Evidence Pools Embedding Neighbor
• Validating Political Position Predictions of Arguments Embedding Neighbor
• PonderLM-2: Pretraining LLM with Latent Thoughts in Continuous Space Embedding Neighbor
• A Scalable Framework for Evaluating Health Language Models Embedding Neighbor
• DySCO: Dynamic Attention-Scaling Decoding for Long-Context LMs Embedding Neighbor
• Unlocking Reasoning Capability on Machine Translation in Large Language Models Embedding Neighbor
• Benchmarking Large Language Models on Answering and Explaining Challenging Medical Questions Embedding Neighbor
• ConformalNL2LTL: Translating Natural Language Instructions into Temporal Logic Formulas with Conformal Correctness Guarantees Tag Overlap
• CORE: Measuring Multi-Agent LLM Interaction Quality under Game-Theoretic Pressures Tag Overlap
• Aladdin-FTI @ AMIYA Three Wishes for Arabic NLP: Fidelity, Diglossia, and Multidialectal Generation Tag Overlap