Spectral Probing

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review


Linguistic information is encoded at varying timescales (subwords, phrases, etc.) and communicative levels, such as syntax and semantics. Contextualized embeddings have analogously been found to capture these phenomena at distinctive layers and frequencies. Leveraging these findings, we develop a fully learnable frequency filter to identify spectral profiles for any given task. It enables vastly more granular analyses than prior handcrafted filters, and improves on efficiency. After demonstrating the informativeness of spectral probing over manual filters in a monolingual setting, we investigate its multilingual characteristics across seven diverse NLP tasks in six languages. Our analyses identify distinctive spectral profiles which quantify cross-task similarity in a linguistically intuitive manner, while remaining consistent across languages—highlighting their potential as robust, lightweight task descriptors.
Original languageEnglish
Title of host publicationProceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Place of PublicationAbu Dhabi, United Arab Emirates
PublisherAssociation for Computational Linguistics
Publication dateDec 2022
Publication statusPublished - Dec 2022


Dive into the research topics of 'Spectral Probing'. Together they form a unique fingerprint.

Cite this