Nicolas Hervé

Phuong-Hang Le, Valentin Pelloin, Arnault Chatelain et al.

We release Pantagruel models, a new family of self-supervised encoder models for French text and speech. Instead of predicting modality-tailored targets such as textual tokens or speech units, Pantagruel learns contextualized target representations in the feature space, allowing modality-specific encoders to capture linguistic and acoustic regularities more effectively. Separate models are pre-trained on large-scale French corpora, including Wikipedia, OSCAR and CroissantLLM for text, together with MultilingualLibriSpeech, LeBenchmark, and INA-100k for speech. INA-100k is a newly introduced 100,000-hour corpus of French audio derived from the archives of the Institut National de l'Audiovisuel (INA), the national repository of French radio and television broadcasts, providing highly diverse audio data. We evaluate Pantagruel across a broad range of downstream tasks spanning both modalities, including those from the standard French benchmarks such as FLUE or LeBenchmark. Across these tasks, Pantagruel models show competitive or superior performance compared to strong French baselines such as CamemBERT, FlauBERT, and LeBenchmark2.0, while maintaining a shared architecture that can seamlessly handle either speech or text inputs. These results confirm the effectiveness of feature-space self-supervised objectives for French representation learning and highlight Pantagruel as a robust foundation for multimodal speech-text understanding.

Valentin Pelloin, Lena Dodson, Émile Chapuis et al.

This paper introduces a computational framework designed to delineate gender distribution biases in topics covered by French TV and radio news. We transcribe a dataset of 11.7k hours, broadcasted in 2023 on 21 French channels. A Large Language Model (LLM) is used in few-shot conversation mode to obtain a topic classification on those transcriptions. Using the generated LLM annotations, we explore the finetuning of a specialized smaller classification model, to reduce the computational cost. To evaluate the performances of these models, we construct and annotate a dataset of 804 dialogues. This dataset is made available free of charge for research purposes. We show that women are notably underrepresented in subjects such as sports, politics and conflicts. Conversely, on topics such as weather, commercials and health, women have more speaking time than their overall average across all subjects. We also observe representations differences between private and public service channels.

Arnault Chatelain, Étienne Ollion, Qianwen Guan et al.

This position paper argues that the under-representation of social science tasks in contemporary LLM benchmarks limits advances in both LLM evaluation and social scientific inquiry. Benchmarks -- standardized tools for assessing computational systems -- are pivotal in the development of artificial intelligence (AI), including large language models (LLMs). Benchmarks do more than measure progress -- they actively structure it, shaping reputations, research agendas, and commercial outcomes. Despite this central role, the social sciences are largely absent from mainstream evaluation frameworks, even though scholars in these fields generate dozens of rigorously annotated, context-sensitive datasets each year. Integrating this work into benchmark design could significantly improve the generalization and robustness of AI models. In turn, models trained on social scientific tasks would likely yield better performance on classic and contemporary tasks in disciplines as diverse as history, sociology, political science or economics. This is all the more pressing as these disciplines are quickly turning to LLMs for assistance. To address this gap, we introduce BenCSSmark, a benchmark composed of datasets annotated by computational social scientists. By integrating social scientific perspectives into benchmarking, BenCSSmark seeks to promote more robust, transparent, and socially relevant AI systems and to foster efficient collaboration.

Béatrice Mazoyer, Nicolas Hervé, Céline Hudelot et al.

In this work, we evaluate the performance of recent text embeddings for the automatic detection of events in a stream of tweets. We model this task as a dynamic clustering problem.Our experiments are conducted on a publicly available corpus of tweets in English and on a similar dataset in French annotated by our team. We show that recent techniques based on deep neural networks (ELMo, Universal Sentence Encoder, BERT, SBERT), although promising on many applications, are not very suitable for this task. We also experiment with different types of fine-tuning to improve these results on French data. Finally, we propose a detailed analysis of the results obtained, showing the superiority of tf-idf approaches for this task.