António Loison, Quentin Macé, Antoine Edy, Victor Xing, Tom Balough, Gabriel de Souza P. Moreira, Bo Liu, Manuel Faysse, Celine Hudelot, Gautier Viaud. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Retrieval-Augmented Generation (RAG) has emerged as a common paradigm to use Large Language Models (LLMs) alongside private and up-to-date knowledge bases. In this work, we address the challenges of using LLM-as-a-Judge when evaluating grounded answers generated by RAG systems. To assess the calibration and discrimination capabilities of judge models, we identify 7 generator failure modes and introduce GroUSE (Grounded QA Unitary Scoring of Evaluators), a meta-evaluation benchmark of 144 unit tests. This benchmark reveals that existing automated RAG evaluation frameworks often overlook important failure modes, even when using GPT-4 as a judge. To improve on the current design of automated RAG evaluation frameworks, we propose a novel pipeline and find that while closed models perform well on GroUSE, state-of-the-art open-source judges do not generalize to our proposed criteria, despite strong correlation with GPT-4’s judgement. Our findings suggest that correlation with GPT-4 is an incomplete proxy for the practical performance of judge models and should be supplemented with evaluations on unit tests for precise failure mode detection. We further show that finetuning Llama-3 on GPT-4’s reasoning traces significantly boosts its evaluation capabilities, improving upon both correlation with GPT-4’s evaluations and calibration on reference situations

A limitation of modern document retrieval embedding methods is that they typically encode passages (chunks) from the same documents independently, often overlooking crucial contextual information from the rest of the document that could greatly improve individual chunk representations.In this work, we introduce ConTEB (Contextaware Text Embedding Benchmark), a benchmark designed to evaluate retrieval models on their ability to leverage document-wide context.Our results show that state-of-the-art embedding models struggle in retrieval scenarios where context is required.To address this limitation, we propose InSeNT (In-sequence Negative Training), a novel contrastive posttraining approach which combined with late chunking pooling enhances contextual representation learning while preserving computational efficiency.Our method significantly improves retrieval quality on ConTEB without sacrificing base model performance.We further find chunks embedded with our method are more robust to suboptimal chunking strategies and larger retrieval corpus sizes.We opensource all artifacts at https://github.com/ illuin-tech/contextual-embeddings.

Documents are visually rich structures that convey information through text, but also figures, page layouts, tables, or even fonts. Since modern retrieval systems mainly rely on the textual information they extract from document pages to index documents -often through lengthy and brittle processes-, they struggle to exploit key visual cues efficiently. This limits their capabilities in many practical document retrieval applications such as Retrieval Augmented Generation (RAG). To benchmark current systems on visually rich document retrieval, we introduce the Visual Document Retrieval Benchmark $\textit{ViDoRe}$, composed of various page-level retrieval tasks spanning multiple domains, languages, and practical settings. The inherent complexity and performance shortcomings of modern systems motivate a new concept; doing document retrieval by directly embedding the images of the document pages. We release $\textit{ColPali}$, a Vision Language Model trained to produce high-quality multi-vector embeddings from images of document pages. Combined with a late interaction matching mechanism, $\textit{ColPali}$ largely outperforms modern document retrieval pipelines while being drastically simpler, faster and end-to-end trainable. We release models, data, code and benchmarks under open licenses at https://hf.co/vidore.

Instruction Fine-Tuning (IFT) is a powerful paradigm that strengthens the zero-shot capabilities of Large Language Models (LLMs), but in doing so induces new evaluation metric requirements.We show LLM-based metrics to be well adapted to these requirements, and leverage them to conduct an investigation of taskspecialization strategies, quantifying the tradeoffs that emerge in practical industrial settings.Our findings offer practitioners actionable insights for real-world IFT model deployment.

Question Answering, including Reading Comprehension, is one of the NLP research areas that has seen significant scientific breakthroughs over the past few years, thanks to the concomitant advances in Language Modeling. Most of these breakthroughs, however, are centered on the English language. In 2020, as a first strong initiative to bridge the gap to the French language, Illuin Technology introduced FQuAD1.1, a French Native Reading Comprehension dataset composed of 60,000+ questions and answers samples extracted from Wikipedia articles. Nonetheless, Question Answering models trained on this dataset have a major drawback: they are not able to predict when a given question has no answer in the paragraph of interest, therefore making unreliable predictions in various industrial use-cases. We introduce FQuAD2.0, which extends FQuAD with 17,000+ unanswerable questions, annotated adversarially, in order to be similar to answerable ones. This new dataset, comprising a total of almost 80,000 questions, makes it possible to train French Question Answering models with the ability of distinguishing unanswerable questions from answerable ones. We benchmark several models with this dataset: our best model, a fine-tuned CamemBERT-large, achieves a F1 score of 82.3% on this classification task, and a F1 score of 83% on the Reading Comprehension task.

The integration of several sources of data for the identification of subtypes of diseases has gained attention over the past few years. The heterogeneity and the high dimensions of the data sets calls for an adequate representation of the data. We summarize the field of representation learning for the multi-omics clustering problem and we investigate several techniques to learn relevant combined representations, using methods from group factor analysis (PCA, MFA and extensions) and from machine learning with autoencoders. We highlight the importance of appropriately designing and training the latter, notably with a novel combination of a disjointed deep autoencoder (DDAE) architecture and a layer-wise reconstruction loss. These different representations can then be clustered to identify biologically meaningful clusters of patients. We provide a unifying framework for model comparison between statistical and deep learning approaches with the introduction of a new weighted internal clustering index that evaluates how well the clustering information is retained from each source, favoring contributions from all data sets. We apply our methodology to two case studies for which previous works of integrative clustering exist, TCGA Breast Cancer and TARGET Neuroblastoma, and show how our method can yield good and well-balanced clusters across the different data sources.

Recently, due to the ubiquity and supremacy of E-recruitment platforms, job recommender systems have been largely studied. In this paper, we tackle the next job application problem, which has many practical applications. In particular, we propose to leverage next-item recommendation approaches to consider better the job seeker’s career preference to discover the next relevant job postings (referred to jobs for short) they might apply for. Our proposed model, named Personalized-Attention Next-Application Prediction (PANAP), is composed of three modules. The first module learns job representations from textual content and metadata attributes in an unsupervised way. The second module learns job seeker representations. It includes a personalized-attention mechanism that can adapt the importance of each job in the learned career preference representation to the specific job seeker’s profile. The attention mechanism also brings some interpretability to learned representations. Then, the third module models the Next-Application Prediction task as a top-K search process based on the similarity of representations. In addition, the geographic location is an essential factor that affects the preferences of job seekers in the recruitment domain. Therefore, we explore the influence of geographic location on the model performance from the perspective of negative sampling strategies. Experiments on the public CareerBuilder12 dataset show the interest in our approach.