Specialised translation relies on the use of documentary and terminological resources, including corpora. These resources are particularly useful for terminology. However, their compilation and exploitation have several limitations: they require time, technical skills and access to data that can be difficult to collect. This study examines the extent to which LLMs can assist specialised translators in finding equivalents from English to French. We evaluate four proprietary models, GPT-4o, GPT-5.2, Claude Sonnet 4.5 and DeepSeek, in two specialised domains, Earth, Environmental and Planetary Sciences (EEPS) and Natural Language Processing (NLP). The experiment is based on 80 terms per domain and compares two prompting strategies: a terminology and a translation mode. The results highlight clear differences between models, prompting strategies and, to a lesser extent, domains. Claude Sonnet 4.5 achieves the best results in the most favourable configuration, while DeepSeek stands out for its greater stability. Analysis of confidence estimates also shows that they are only a partial indicator of terminological accuracy. Overall, the findings suggest that LLMs can be useful tools for specialised translators, but cannot, at this stage, replace specialised corpora. This research therefore paves the way for future work on the real practical usefulness of LLMs for specialised translators in work and educational contexts.

Unlike text, speech conveys information about the speaker, such as gender, through acoustic cues like pitch. This gives rise to modality-specific bias concerns. For example, in speech translation (ST), when translating from languages with notional gender, such as English, into languages where gender-ambiguous terms referring to the speaker are assigned grammatical gender, the speaker’s vocal characteristics may play a role in gender assignment. This risks misgendering speakers—whether through masculine defaults or vocal-based assumptions—yet how ST models make these decisions remains poorly understood. We investigate the mechanisms ST models use to assign gender to speaker-referring terms across three language pairs (en→es/fr/it). To do so, we examine how training data patterns, internal language model (ILM) biases, and acoustic information interact. We find that models do not simply replicate term-specific gender associations from training data, but learn broader patterns of masculine prevalence. While the ILM exhibits strong masculine bias, models can override these preferences based on acoustic input. Using contrastive feature attribution on spectrograms, we reveal that the model with higher gender accuracy relies on a previously unknown mechanism: using first-person pronouns to link gendered terms back to the speaker, accessing gender information distributed across the frequency spectrum rather than concentrated in pitch.

Many studies have explored when and how LLMs learn to use specific words, primarily by examining their learning curves. While these curves capture a model’s capacity to use words correctly in context, they often neglect the equally important skill of avoiding incorrect usage. In this paper, we introduce a new metric, anti-surprisal, which measures a model’s capacity to refrain from using words in inappropriate or unexpected contexts. By examining both correct usage and error avoidance, we offer a more comprehensive perspective on the learning dynamics of LLMs.

This study investigates the capabilities of large language models (LLMs), specifically ChatGPT, in annotating MT outputs based on an error typology. In contrast to previous work focusing mainly on general language, we explore ChatGPT’s ability to identify and categorise errors in specialised translations. By testing two different prompts and based on a customised error typology, we compare ChatGPT annotations with human expert evaluations of translations produced by DeepL and ChatGPT itself. The results show that, for translations generated by DeepL, recall and precision are quite high. However, the degree of accuracy in error categorisation depends on the prompt’s specific features and its level of detail, ChatGPT performing very well with a detailed prompt. When evaluating its own translations, ChatGPT achieves significantly poorer results, revealing limitations with self-assessment. These results highlight both the potential and the limitations of LLMs for translation evaluation, particularly in specialised domains. Our experiments pave the way for future research on open-source LLMs, which could produce annotations of comparable or even higher quality. In the future, we also aim to test the practical effectiveness of this automated evaluation in the context of translation training, particularly by optimising the process of human evaluation by teachers and by exploring the impact of annotations by LLMs on students’ post-editing and translation learning.

International audience

Numerous studies have demonstrated the ability of neural language models to learn various linguistic properties without direct supervision.This work takes an initial step towards exploring the less researched topic of how neural models discover linguistic properties of words, such as gender, as well as the rules governing their usage.We propose to use an artificial corpus generated by a PCFG based on French to precisely control the gender distribution in the training data and determine under which conditions a model correctly captures gender information or, on the contrary, appears gender-biased.

ISSN: 2958-1796

Abstract Many studies have shown that transformers are able to predict subject-verb agreement, demonstrating their ability to uncover an abstract representation of the sentence in an unsupervised way. Recently, Li et al. (2021) found that transformers were also able to predict the object-past participle agreement in French, the modeling of which in formal grammar is fundamentally different from that of subject-verb agreement and relies on a movement and an anaphora resolution. To better understand transformers’ internal working, we propose to contrast how they handle these two kinds of agreement. Using probing and counterfactual analysis methods, our experiments on French agreements show that (i) the agreement task suffers from several confounders that partially question the conclusions drawn so far and (ii) transformers handle subject-verb and object-past participle agreements in a way that is consistent with their modeling in theoretical linguistics.

International audience

International audience

We evaluate the ability of Bert embeddings to represent tense information, taking French and Chinese as a case study.In French, the tense information is expressed by verb morphology and can be captured by simple surface information.On the contrary, tense interpretation in Chinese is driven by abstract, lexical, syntactic and even pragmatic information.We show that while French tenses can easily be predicted from sentence representations, results drop sharply for Chinese, which suggests that Bert is more likely to memorize shallow patterns from the training data rather than uncover abstract properties.

Many recent works have demonstrated that unsupervised sentence representations of neural networks encode syntactic information by observing that neural language models are able to predict the agreement between a verb and its subject.We take a critical look at this line of research by showing that it is possible to achieve high accuracy on this agreement task with simple surface heuristics, indicating a possible flaw in our assessment of neural networks' syntactic ability.Our fine-grained analyses of results on the long-range French objectverb agreement show that contrary to LSTMs, Transformers are able to capture a non-trivial amount of grammatical structure.

Not all dependencies are equal when training a dependency parser: some are straightforward enough to be learned with only a sample of data, others embed more complexity. This work introduces a series of metrics to quantify those differences, and thereby to expose the shortcomings of various parsing algorithms and strategies. Apart from a more thorough comparison of parsing systems, these new tools also prove useful for characterizing the information conveyed by cross-lingual parsers, in a quantitative but still interpretable way.

International audience

This paper studies cross-lingual transfer for dependency parsing, focusing on very low-resource settings where delexicalized transfer is the only fully automatic option. We show how to boost parsing performance by rewriting the source sentences so as to better match the linguistic regularities of the target language. We contrast a data-driven approach with an approach relying on linguistically motivated rules automatically extracted from the World Atlas of Language Structures. Our findings are backed up by experiments involving 40 languages. They show that both approaches greatly outperform the baseline, the knowledge-driven method yielding the best accuracies, with average improvements of +2.9 UAS, and up to +90 UAS (absolute) on some frequent PoS configurations.

Ophélie Lacroix, Lauriane Aufrant, Guillaume Wisniewski, François Yvon. Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2016.

International audience

International audience

Many problems in natural language processing, information extraction or bioinformatics consist in predicting a label for each element of a sequence of observations. The sequence of labels generally presents multiple dependencies that restrict the possible labels the elements can take. Therefore, relations between labels intuitively provide information valuable for the prediction. Several approaches have been proposed to take advantage of this additional information. However, experimental results show that taking relations into account does not always improve prediction performances, while it significantly increases the computational cost of both learning and prediction. In this work, we aim at both explaining these surprising results and proposing a simple but computationally efficient approach for labeling sequences.