Several methodologies have recently been proposed to evaluate the ability of Pretrained Language Models (PLMs) to interpret negation. In this article, we build on Gubelmann and Handschuh (2022), which studies the modification of PLMs’ predictions as a function of the polarity of inputs, in English. Crucially, this test uses “self-contained” inputs ending with a masked position: depending on the polarity of a verb in the input, a particular token is either semantically ruled out or allowed at the masked position. By replicating Gubelmann and Handschuh (2022) experiments, we have uncovered flaws that weaken the conclusions that can be drawn from this test. We thus propose an improved version, the Self-Contained Neg Test, which is more controlled, more systematic, and entirely based on examples forming minimal pairs varying only in the presence or absence of verbal negation in English. When applying our test to the roberta and bert base and large models, we show that only roberta-large shows trends that match the expectations, while bert-base is mostly insensitive to negation. For all the tested models though, in a significant number of test instances the top-1 prediction remains the token that is semantically forbidden by the context, which shows how much room for improvement remains for a proper treatment of the negation phenomenon.

Polysemies, or “colexifications”, are of great interest in cognitive and historical linguistics, since meanings that are frequently expressed by the same lexeme are likely to be conceptually similar, and lie along a common pathway of semantic change. We argue that these types of inferences can be more reliably drawn from polysemies of cognate sets (which we call “dialexifications”) than from polysemies of lexemes. After giving a precise definition of dialexification, we introduce Evosem, a cross-linguistic database of etymologies scraped from several online sources. Based on this database, we measure for each pair of senses how many cognate sets include them both — i.e. how often this pair of senses is “dialexified”. This allows us to construct a weighted dialexification graph for any set of senses, indicating the conceptual and historical closeness of each pair. We also present an online interface for browsing our database, including graphs and interactive tables. We then discuss potential applications to NLP tasks and to linguistic research.

Contradictory results about the encoding of the semantic impact of negation in pretrained language models (PLMs) have been drawn recently (e.g. Kassner and Schütze (2020); Gubelmann and Handschuh (2022)).In this paper we focus rather on the way PLMs encode negation and its formal impact, through the phenomenon of the Negative Polarity Item (NPI) licensing in English.More precisely, we use probes to identify which contextual representations best encode 1) the presence of negation in a sentence, and 2) the polarity of a neighboring masked polarity item. We find that contextual representations of tokens inside the negation scope do allow for (i) a better prediction of the presence of “not” compared to those outside the scope and (ii) a better prediction of the right polarity of a masked polarity item licensed by “not”, although the magnitude of the difference varies from PLM to PLM. Importantly, in both cases the trend holds even when controlling for distance to “not”.This tends to indicate that the embeddings of these models do reflect the notion of negation scope, and do encode the impact of negation on NPI licensing. Yet, further control experiments reveal that the presence of other lexical items is also better captured when using the contextual representation of a token within the same syntactic clause than outside from it, suggesting that PLMs simply capture the more general notion of syntactic clause.

Different algorithms have been proposed to detect semantic shifts (changes in a word meaning over time) in a diachronic corpus. Yet, and somehow surprisingly, no reference corpus has been designed so far to evaluate them, leaving researchers to fallback to troublesome evaluation strategies. In this work, we introduce a methodology for the construction of a reference dataset for the evaluation of semantic shift detection, that is, a list of words where we know for sure whether they present a word meaning change over a period of interest. We leverage a state-of-the-art word-sense disambiguation model to associate a date of first appearance to all the senses of a word. Significant changes in sense distributions as well as clear stability are detected and the resulting words are inspected by experts using a dedicated interface before populating a reference dataset. As a proof of concept, we apply this methodology to a corpus of newspapers from Quebec covering the whole 20th century. We manually verified a subset of candidates, leading to QC-FR-Diac-V1.0, a corpus of 151 words allowing one to evaluate the identification of semantic shifts in French between 1910 and 1990.