Francisco Casacuberta

Blai Puchol, Sergio Gómez González, Miguel Domingo et al.

This work presents a comparative evaluation of machine translation systems applied to images containing textual information, a task that lies at the intersection of computer vision and natural language processing. The study compares three main paradigms: modular pipelines that separate text detection, recognition, and translation; multi-modal large language models (MLLMs) capable of processing both image and text jointly; and an end-to-end model, Translatotron-V, which directly generates translated images. The modular systems employ state-of-the-art OCR (docTR) combined with multilingual LLMs such as Llama and EuroLLM, while the evaluated MLLMs include different configurations of Gemini 2.5. Experiments were conducted on parallel multilingual datasets covering multiple language pairs, with evaluation based on BLEU, chrF, and TER metrics. The results show that modular pipelines outperform the end-to-end approach, while MLLMs achieve the best overall performance, demonstrating superior flexibility and contextual understanding. These findings underscore the effectiveness of multi-modal reasoning for image-to-text translation and provide a solid foundation for future research on integrating visual understanding and language generation in multilingual settings.

Salvador Carrión, Francisco Casacuberta

We present AutoNMT, a framework to streamline the research of seq-to-seq models by automating the data pipeline (i.e., file management, data preprocessing, and exploratory analysis), automating experimentation in a toolkit-agnostic manner, which allows users to use either their own models or existing seq-to-seq toolkits such as Fairseq or OpenNMT, and finally, automating the report generation (plots and summaries). Furthermore, this library comes with its own seq-to-seq toolkit so that users can easily customize it for non-standard tasks.

Francisco Casacuberta, Alexandru Ceausu, Khalid Choukri et al.

This work presents the results of the machine translation (MT) task from the Covid-19 MLIA @ Eval initiative, a community effort to improve the generation of MT systems focused on the current Covid-19 crisis. Nine teams took part in this event, which was divided in two rounds and involved seven different language pairs. Two different scenarios were considered: one in which only the provided data was allowed, and a second one in which the use of external resources was allowed. Overall, best approaches were based on multilingual models and transfer learning, with an emphasis on the importance of applying a cleaning process to the training data.

Sergio Gómez González, Miguel Domingo, Francisco Casacuberta

Comparative evaluation of several systems is a recurrent task in researching. It is a key step before deciding which system to use for our work, or, once our research has been conducted, to demonstrate the potential of the resulting model. Furthermore, it is the main task of competitive, public challenges evaluation. Our proposed software (DEEP) automates both the execution and scoring of machine translation and optical character recognition models. Furthermore, it is easily extensible to other tasks. DEEP is prepared to receive dockerized systems, run them (extracting information at that same time), and assess hypothesis against some references. With this approach, evaluators can achieve a better understanding of the performance of each model. Moreover, the software uses a clustering algorithm based on a statistical analysis of the significance of the results yielded by each model, according to the evaluation metrics. As a result, evaluators are able to identify clusters of performance among the swarm of proposals and have a better understanding of the significance of their differences. Additionally, we offer a visualization web-app to ensure that the results can be adequately understood and interpreted. Finally, we present an exemplary case of use of DEEP.

Angel Navarro, Francisco Casacuberta

Pre-trained large language models (LLM) are starting to be widely used in many applications. In this work, we explore the use of these models in interactive machine translation (IMT) environments. In particular, we have chosen mBART (multilingual Bidirectional and Auto-Regressive Transformer) and mT5 (multilingual Text-to-Text Transfer Transformer) as the LLMs to perform our experiments. The system generates perfect translations interactively using the feedback provided by the user at each iteration. The Neural Machine Translation (NMT) model generates a preliminary hypothesis with the feedback, and the user validates new correct segments and performs a word correction--repeating the process until the sentence is correctly translated. We compared the performance of mBART, mT5, and a state-of-the-art (SoTA) machine translation model on a benchmark dataset regarding user effort, Word Stroke Ratio (WSR), Key Stroke Ratio (KSR), and Mouse Action Ratio (MAR). The experimental results indicate that mBART performed comparably with SoTA models, suggesting that it is a viable option for this field of IMT. The implications of this finding extend to the development of new machine translation models for interactive environments, as it indicates that some novel pre-trained models exhibit SoTA performance in this domain, highlighting the potential benefits of adapting these models to specific needs.

Miguel Domingo, Francisco Casacuberta

We present our demonstration of two machine translation applications to historical documents. The first task consists in generating a new version of a historical document, written in the modern version of its original language. The second application is limited to a document's orthography. It adapts the document's spelling to modern standards in order to achieve an orthography consistency and accounting for the lack of spelling conventions. We followed an interactive, adaptive framework that allows the user to introduce corrections to the system's hypothesis. The system reacts to these corrections by generating a new hypothesis that takes them into account. Once the user is satisfied with the system's hypothesis and validates it, the system adapts its model following an online learning strategy. This system is implemented following a client-server architecture. We developed a website which communicates with the neural models. All code is open-source and publicly available. The demonstration is hosted at http://demosmt.prhlt.upv.es/mthd/.

Álvaro Peris, Francisco Casacuberta

We present a demonstration of a neural interactive-predictive system for tackling multimodal sequence to sequence tasks. The system generates text predictions to different sequence to sequence tasks: machine translation, image and video captioning. These predictions are revised by a human agent, who introduces corrections in the form of characters. The system reacts to each correction, providing alternative hypotheses, compelling with the feedback provided by the user. The final objective is to reduce the human effort required during this correction process. This system is implemented following a client-server architecture. For accessing the system, we developed a website, which communicates with the neural model, hosted in a local server. From this website, the different tasks can be tackled following the interactive-predictive framework. We open-source all the code developed for building this system. The demonstration in hosted in http://casmacat.prhlt.upv.es/interactive-seq2seq.

Salvador Carrión, Francisco Casacuberta

Continual learning in Neural Machine Translation (NMT) faces the dual challenges of catastrophic forgetting and the high computational cost of retraining. This study establishes Low-Rank Adaptation (LoRA) as a parameter-efficient framework to address these challenges in dedicated NMT architectures. We first demonstrate that LoRA-based fine-tuning adapts NMT models to new languages and domains with performance on par with full-parameter techniques, while utilizing only a fraction of the parameter space. Second, we propose an interactive adaptation method using a calibrated linear combination of LoRA modules. This approach functions as a gate-free mixture of experts, enabling real-time, user-controllable adjustments to domain and style without retraining. Finally, to mitigate catastrophic forgetting, we introduce a novel gradient-based regularization strategy specifically designed for low-rank decomposition matrices. Unlike methods that regularize the full parameter set, our approach weights the penalty on the low-rank updates using historical gradient information. Experimental results indicate that this strategy efficiently preserves prior domain knowledge while facilitating the acquisition of new tasks, offering a scalable paradigm for interactive and continual NMT.

Miguel Domingo, Francisco Casacuberta

The absence of standardized spelling conventions and the organic evolution of human language present an inherent linguistic challenge within historical documents, a longstanding concern for scholars in the humanities. Addressing this issue, spelling normalization endeavors to align a document's orthography with contemporary standards. In this study, we propose two new approaches based on large language models: one of which has been trained without a supervised training, and a second one which has been trained for machine translation. Our evaluation spans multiple datasets encompassing diverse languages and historical periods, leading us to the conclusion that while both of them yielded encouraging results, statistical machine translation still seems to be the most suitable technology for this task.

Miguel Domingo, Francisco Casacuberta

Due to the nature of human language, historical documents are hard to comprehend by contemporary people. This limits their accessibility to scholars specialized in the time period in which the documents were written. Modernization aims at breaking this language barrier by generating a new version of a historical document, written in the modern version of the document's original language. However, while it is able to increase the document's comprehension, modernization is still far from producing an error-free version. In this work, we propose a collaborative framework in which a scholar can work together with the machine to generate the new version. We tested our approach on a simulated environment, achieving significant reductions of the human effort needed to produce the modernized version of the document.

Miguel Domingo, Francisco Casacuberta

Accessibility to historical documents is mostly limited to scholars. This is due to the language barrier inherent in human language and the linguistic properties of these documents. Given a historical document, modernization aims to generate a new version of it, written in the modern version of the document's language. Its goal is to tackle the language barrier, decreasing the comprehension difficulty and making historical documents accessible to a broader audience. In this work, we proposed a new neural machine translation approach that profits from modern documents to enrich its systems. We tested this approach with both automatic and human evaluation, and conducted a user study. Results showed that modernization is successfully reaching its goal, although it still has room for improvement.

Miguel Domingo, Mercedes García-Martínez, Amando Estela et al.

We introduce a demonstration of our system, which implements online learning for neural machine translation in a production environment. These techniques allow the system to continuously learn from the corrections provided by the translators. We implemented an end-to-end platform integrating our machine translation servers to one of the most common user interfaces for professional translators: SDL Trados Studio. Our objective was to save post-editing effort as the machine is continuously learning from human choices and adapting the models to a specific domain or user style.

Miguel Domingo, Mercedes García-Martínez, Álvaro Peris et al.

A common use of machine translation in the industry is providing initial translation hypotheses, which are later supervised and post-edited by a human expert. During this revision process, new bilingual data are continuously generated. Machine translation systems can benefit from these new data, incrementally updating the underlying models under an online learning paradigm. We conducted a user study on this scenario, for a neural machine translation system. The experimentation was carried out by professional translators, with a vast experience in machine translation post-editing. The results showed a reduction in the required amount of human effort needed when post-editing the outputs of the system, improvements in the translation quality and a positive perception of the adaptive system by the users.

Álvaro Peris, Francisco Casacuberta

Despite the advances achieved by neural models in sequence to sequence learning, exploited in a variety of tasks, they still make errors. In many use cases, these are corrected by a human expert in a posterior revision process. The interactive-predictive framework aims to minimize the human effort spent on this process by considering partial corrections for iteratively refining the hypothesis. In this work, we generalize the interactive-predictive approach, typically applied in to machine translation field, to tackle other multimodal problems namely, image and video captioning. We study the application of this framework to multimodal neural sequence to sequence models. We show that, following this framework, we approximately halve the effort spent for correcting the outputs generated by the automatic systems. Moreover, we deploy our systems in a publicly accessible demonstration, that allows to better understand the behavior of the interactive-predictive framework.

Miguel Domingo, Mercedes Garcıa-Martınez, Alexandre Helle et al.

Tokenization or segmentation is a wide concept that covers simple processes such as separating punctuation from words, or more sophisticated processes such as applying morphological knowledge. Neural Machine Translation (NMT) requires a limited-size vocabulary for computational cost and enough examples to estimate word embeddings. Separating punctuation and splitting tokens into words or subwords has proven to be helpful to reduce vocabulary and increase the number of examples of each word, improving the translation quality. Tokenization is more challenging when dealing with languages with no separator between words. In order to assess the impact of the tokenization in the quality of the final translation on NMT, we experimented on five tokenizers over ten language pairs. We reached the conclusion that the tokenization significantly affects the final translation quality and that the best tokenizer differs for different language pairs.

Álvaro Peris, Francisco Casacuberta

We study the application of active learning techniques to the translation of unbounded data streams via interactive neural machine translation. The main idea is to select, from an unbounded stream of source sentences, those worth to be supervised by a human agent. The user will interactively translate those samples. Once validated, these data is useful for adapting the neural machine translation model. We propose two novel methods for selecting the samples to be validated. We exploit the information from the attention mechanism of a neural machine translation system. Our experiments show that the inclusion of active learning techniques into this pipeline allows to reduce the effort required during the process, while increasing the quality of the translation system. Moreover, it enables to balance the human effort required for achieving a certain translation quality. Moreover, our neural system outperforms classical approaches by a large margin.

Álvaro Peris, Francisco Casacuberta

We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and Tensorflow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering.

Álvaro Peris, Francisco Casacuberta

Neural machine translation systems require large amounts of training data and resources. Even with this, the quality of the translations may be insufficient for some users or domains. In such cases, the output of the system must be revised by a human agent. This can be done in a post-editing stage or following an interactive machine translation protocol. We explore the incremental update of neural machine translation systems during the post-editing or interactive translation processes. Such modifications aim to incorporate the new knowledge, from the edited sentences, into the translation system. Updates to the model are performed on-the-fly, as sentences are corrected, via online learning techniques. In addition, we implement a novel interactive, adaptive system, able to react to single-character interactions. This system greatly reduces the human effort required for obtaining high-quality translations. In order to stress our proposals, we conduct exhaustive experiments varying the amount and type of data available for training. Results show that online learning effectively achieves the objective of reducing the human effort required during the post-editing or the interactive machine translation stages. Moreover, these adaptive systems also perform well in scenarios with scarce resources. We show that a neural machine translation system can be rapidly adapted to a specific domain, exclusively by means of online learning techniques.

Álvaro Peris, Luis Cebrián, Francisco Casacuberta

Neural machine translation has meant a revolution of the field. Nevertheless, post-editing the outputs of the system is mandatory for tasks requiring high translation quality. Post-editing offers a unique opportunity for improving neural machine translation systems, using online learning techniques and treating the post-edited translations as new, fresh training data. We review classical learning methods and propose a new optimization algorithm. We thoroughly compare online learning algorithms in a post-editing scenario. Results show significant improvements in translation quality and effort reduction.

Marc Bolaños, Álvaro Peris, Francisco Casacuberta et al.

Egocentric vision consists in acquiring images along the day from a first person point-of-view using wearable cameras. The automatic analysis of this information allows to discover daily patterns for improving the quality of life of the user. A natural topic that arises in egocentric vision is storytelling, that is, how to understand and tell the story relying behind the pictures. In this paper, we tackle storytelling as an egocentric sequences description problem. We propose a novel methodology that exploits information from temporally neighboring events, matching precisely the nature of egocentric sequences. Furthermore, we present a new method for multimodal data fusion consisting on a multi-input attention recurrent network. We also publish the first dataset for egocentric image sequences description, consisting of 1,339 events with 3,991 descriptions, from 55 days acquired by 11 people. Furthermore, we prove that our proposal outperforms classical attentional encoder-decoder methods for video description.

Álvaro Peris, Mara Chinea-Rios, Francisco Casacuberta

We address the data selection problem in statistical machine translation (SMT) as a classification task. The new data selection method is based on a neural network classifier. We present a new method description and empirical results proving that our data selection method provides better translation quality, compared to a state-of-the-art method (i.e., Cross entropy). Moreover, the empirical results reported are coherent across different language pairs.

Marc Bolaños, Álvaro Peris, Francisco Casacuberta et al.

In this paper, we address the problem of visual question answering by proposing a novel model, called VIBIKNet. Our model is based on integrating Kernelized Convolutional Neural Networks and Long-Short Term Memory units to generate an answer given a question about an image. We prove that VIBIKNet is an optimal trade-off between accuracy and computational load, in terms of memory and time consumption. We validate our method on the VQA challenge dataset and compare it to the top performing methods in order to illustrate its performance and speed.

Álvaro Peris, Marc Bolaños, Petia Radeva et al.

Although traditionally used in the machine translation field, the encoder-decoder framework has been recently applied for the generation of video and image descriptions. The combination of Convolutional and Recurrent Neural Networks in these models has proven to outperform the previous state of the art, obtaining more accurate video descriptions. In this work we propose pushing further this model by introducing two contributions into the encoding stage. First, producing richer image representations by combining object and location information from Convolutional Neural Networks and second, introducing Bidirectional Recurrent Neural Networks for capturing both forward and backward temporal relationships in the input frames.