CL-IMS @ DIACR-Ita: Volente o Nolente: BERT does not outperform SGNS on\n Semantic Change Detection

2020/11/14 by Severin Laicher, Gioia Baldissin, Laicher, Severin +7
Social Sciences · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Language and cultural evolution

paper · pdf · doi:10.48550/arxiv.2011.07247

openalex publication_date 2020/11/14 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28

Abstract

We present the results of our participation in the DIACR-Ita shared task on\nlexical semantic change detection for Italian. We exploit Average Pairwise\nDistance of token-based BERT embeddings between time points and rank 5 (of 8)\nin the official ranking with an accuracy of .72. While we tune parameters on\nthe English data set of SemEval-2020 Task 1 and reach high performance, this\ndoes not translate to the Italian DIACR-Ita data set. Our results show that we\ndo not manage to find robust ways to exploit BERT embeddings in lexical\nsemantic change detection.\n

Related

• OP-IMS @ DIACR-Ita: Back to the Roots: SGNS+OP+CD still rocks Semantic\n Change Detection
• Explaining and Improving BERT Performance on Lexical Semantic Change\n Detection
• UWB @ DIACR-Ita: Lexical Semantic Change Detection with CCA and Orthogonal Transformation
• QMUL-SDS @ DIACR-Ita: Evaluating Unsupervised Diachronic Lexical Semantics Classification in Italian
• NLP-CIC @ DIACR-Ita: POS and Neighbor Based Distributional Models for Lexical Semantic Change in Diachronic Italian Corpora