arXiv subjects

Ricardo Rei

Publications and source records attributed to Ricardo Rei.

At least 19 recordsLinked to original sources

arXiv2026

Move by Move: Measuring and Steering How LLMs Conduct Psychotherapy

Users increasingly turn to large language models for emotional support, yet little is known about how these models actually conduct a psychotherapy interaction. We introduce an ontology of ten therapeutic moves: compact, function-based categories grounded in the MULTI-60 inventory, validated through an annotation campaign with five licensed psychologists, and scaled with a judge-based approach that matches expert agreement. Applying it to real counseling transcripts and model-led sessions, we compare the move distributions between human clinicians and a panel of frontier models. Models over-use inquiry at up to three times the human rate, neglect psychoeducation, and are strongly context-anchored: they carry forward strategies initiated by a human clinician but rarely initiate them themselves. Exposing the ontology as a set of tools roughly halves the mean deviation from the human move distribution and improves turn-level alignment with human therapist by 7-9 percentage points, without any fine-tuning.

cs.CL↗

arXiv2026

IndicQE-APE: A Benchmark for Quality Estimation and Automatic Post-Editing for Indic Languages

Indic quality estimation (QE) and automatic post-editing (APE) data is spread across separate releases, so no single resource supports training and evaluation across tasks and language pairs on one footing. We consolidate the WMT 2020--2024 shared-task lineage with an extended English--Malayalam resource into \indicqe: $126{,}754$ instances over nine directional pairs, with up to four label types aligned on the same segment, a direct assessment, a human post-edit, word-level OK/BAD tags and an error explanation, and a test set stratified over four difficulty axes. On it we benchmark six prompted LLMs and three COMET metrics on segment-level QE, and three systems on APE. Two of the axes are defined partly on the direct assessment and select a compressed slice of it, so each axis is compared against a control drawn from the same language pair with the same score distribution. Only one survives that control: segments whose holistic and token-level quality signals conflict are ranked worse than equally-scored segments of the same language, for all nine systems and all seven pairs that carry the axis. Annotator disagreement, which looks second-hardest without the control, has no effect with it. Few-shot prompting costs every model $\leq$ $3.4$B both correlation and output-format compliance. Within-language accuracy does not make scores comparable across pairs: of the three trained metrics, the one with the best within-language correlation loses most when the pairs are pooled. The benchmark (https://huggingface.co/datasets/surrey-nlp/IndicQE-APE) and code (https://github.com/surrey-nlp/IndicQE-APE) are released.

cs.CL↗

arXiv2026

Risk Governance for Generative AI Mental Health Support: A Multi-Turn Safety Architecture

Large language models (LLMs) are increasingly used for emotional support despite lacking mechanisms to safely govern evolving mental health risk. Existing safety approaches primarily detect risk but rarely shape how models respond as conversational risk unfolds. We developed a model-agnostic safety governance architecture that combines contextual risk detection, reasoning-based verification, and protocol-guided response generation for multi-turn mental health interactions. Synthetic conversations grounded in real-world mental health narratives were used to evaluate the architecture's performance, tested with GPT-5-chat and Qwen3.5-27B, achieving high risk detection performance (specificity: 0.85 (95\%CI: 0.78;0.91), sensitivity: 0.92 (95\%CI: 0.88;0.95)) and increasing clinician-preferred escalation responses by 25.6--59.2pp while preserving rapport and connection. Performance remained stable across conversation length and generalized across both proprietary and open-source models. These findings demonstrate that clinically-grounded safety governance can extend beyond risk detection to improve how LLMs manage evolving mental health risk, providing a scalable framework for safer deployment across models.

cs.AI↗

arXiv2026

Can Vision Language Models Judge Action Quality? An Empirical Evaluation

Action Quality Assessment (AQA) has broad applications in physical therapy, sports coaching, and competitive judging. Although Vision Language Models (VLMs) hold considerable promise for AQA, their actual performance in this domain remains largely uncharacterised. We present a comprehensive evaluation of state-of-the-art VLMs across activity domains (e.g. fitness, figure skating, diving), tasks, representations, and prompting strategies. Baseline results reveal that Gemini 3.1 Pro, Qwen3-VL and InternVL3.5 models perform only marginally above random chance, and although strategies such as incorporation of skeleton information, grounding instructions, reasoning structures and in-context learning lead to isolated gains, none is consistently effective. Analysis of prediction distributions uncovers two systematic biases: a tendency to predict correct execution regardless of visual evidence, and a sensitivity to superficial linguistic framing. Reformulating tasks contrastively to mitigate these biases yields minimal improvement, suggesting that the models' limitations go beyond these biases, pointing to a fundamental difficulty with fine-grained movement quality assessment. Our findings establish a rigorous baseline for future VLM-based AQA research and provide an actionable outline for failure modes requiring mitigation prior to reliable real-world deployment.

cs.CV↗

arXiv2026

Self-Preference Bias in Rubric-Based Evaluation of Large Language Models

LLM-as-a-judge has become the de facto approach for evaluating LLM outputs. However, judges are known to exhibit self-preference bias (SPB): they tend to favor outputs produced by themselves or by models from their own family. This skews evaluations and, thus, hinders model development, especially in settings of recursive self-improvement. We present the first study of SPB in rubric-based evaluation, an increasingly popular benchmarking paradigm where judges issue binary verdicts on individual evaluation criteria, instead of assigning holistic scores or rankings. Using IFEval and LiveCodeBench, benchmarks with programmatically verifiable rubrics, we show that SPB persists even when evaluation criteria are entirely objective: among rubrics where generators fail, judges can be more than 50% more likely to incorrectly mark them as satisfied when the output is their own. We also find that, similarly to other evaluation paradigms, ensembling multiple judges helps mitigate SPB, but without fully eliminating it. On HealthBench, a medical chat benchmark with subjective rubrics, we observe that SPB skews model scores by up to 10 points, a potentially decisive margin when ranking frontier models. We analyze the factors that drive SPB in this setting, finding that negative rubrics and subjective topics like communication and emergency referrals are particularly susceptible.

cs.CL↗

arXiv2026

EuroLLM-22B: Technical Report

This report presents EuroLLM-22B, a large language model trained from scratch to support the needs of European citizens by covering all 24 official European Union languages and 11 additional languages. EuroLLM addresses the issue of European languages being underrepresented and underserved in existing open large language models. We provide a comprehensive overview of EuroLLM-22B's development, including tokenizer design, architectural specifications, data filtering, and training procedures. Across a broad set of multilingual benchmarks, EuroLLM-22B demonstrates strong performance in reasoning, instruction following, and translation, achieving results competitive with models of comparable size. To support future research, we release our base and instruction-tuned models, our multilingual web pretraining data and updated EuroBlocks instruction datasets, as well as our pre-training and evaluation codebases.

cs.CL↗

arXiv2026

MindGuard: Guardrail Classifiers for Multi-Turn Mental Health Support

Large language models are increasingly used for mental health support, yet their conversational coherence alone does not ensure clinical appropriateness. Existing general-purpose safeguards often fail to distinguish between therapeutic disclosures and genuine clinical crises, leading to safety failures. To address this gap, we introduce a clinically grounded risk taxonomy, developed in collaboration with PhD-level psychologists, that identifies actionable harm (e.g., self-harm and harm to others) while preserving space for safe, non-crisis therapeutic content. We release MindGuard-testset, a dataset of real-world multi-turn conversations annotated at the turn level by clinical experts. Using synthetic dialogues generated via a controlled two-agent setup, we train MindGuard, a family of lightweight safety classifiers (with 4B and 8B parameters). Our classifiers reduce false positives at high-recall operating points and, when paired with clinician language models, help achieve lower attack success and harmful engagement rates in adversarial multi-turn interactions compared to general-purpose safeguards. We release all models and human evaluation data.

cs.AI↗

arXiv2025

MindEval: Benchmarking Language Models on Multi-turn Mental Health Support

Demand for mental health support through AI chatbots is surging, though current systems present several limitations, like sycophancy or overvalidation, and reinforcement of maladaptive beliefs. A core obstacle to the creation of better systems is the scarcity of benchmarks that capture the complexity of real therapeutic interactions. Most existing benchmarks either only test clinical knowledge through multiple-choice questions or assess single responses in isolation. To bridge this gap, we present MindEval, a framework designed in collaboration with Ph.D-level Licensed Clinical Psychologists for automatically evaluating language models in realistic, multi-turn mental health therapy conversations. Through patient simulation and automatic evaluation with LLMs, our framework balances resistance to gaming with reproducibility via its fully automated, model-agnostic design. We begin by quantitatively validating the realism of our simulated patients against human-generated text and by demonstrating strong correlations between automatic and human expert judgments. Then, we evaluate 12 state-of-the-art LLMs and show that all models struggle, scoring below 4 out of 6, on average, with particular weaknesses in problematic AI-specific patterns of communication. Notably, reasoning capabilities and model scale do not guarantee better performance, and systems deteriorate with longer interactions or when supporting patients with severe symptoms. We release all code, prompts, and human evaluation data.

cs.CL↗

arXiv2025

Zero-shot Benchmarking: A Framework for Flexible and Scalable Automatic Evaluation of Language Models

As language models improve and become capable of performing more complex tasks across modalities, evaluating them automatically becomes increasingly challenging. Developing strong and robust task-specific automatic metrics gets harder, and human-annotated test sets -- which are expensive to create -- saturate more quickly. A compelling alternative is to design reliable strategies to automate the creation of test data and evaluation, but previous attempts either rely on pre-existing data, or focus solely on individual tasks. We present Zero-shot Benchmarking (ZSB), a framework for creating high-quality benchmarks for any task by leveraging language models for both synthetic test data creation and evaluation. ZSB is simple and flexible: it requires only the creation of a prompt for data generation and one for evaluation; it is scalable to tasks and languages where collecting real-world data is costly or impractical; it is model-agnostic, allowing the creation of increasingly challenging benchmarks as models improve. To assess the effectiveness of our framework, we create benchmarks for five text-only tasks and a multi-modal one: general capabilities in four languages (English, Chinese, French, and Korean), translation, and general vision-language capabilities in English. We then rank a broad range of open and closed systems on our benchmarks. ZSB rankings consistently correlate strongly with human rankings, outperforming widely-adopted standard benchmarks. Through ablations, we find that strong benchmarks can be created with open models, and that judge model size and dataset variety are crucial drivers of performance. We release all our benchmarks, and code to reproduce our experiments and to produce new benchmarks.

cs.CL↗

arXiv2025

M-Prometheus: A Suite of Open Multilingual LLM Judges

The use of language models for automatically evaluating long-form text (LLM-as-a-judge) is becoming increasingly common, yet most LLM judges are optimized exclusively for English, with strategies for enhancing their multilingual evaluation capabilities remaining largely unexplored in the current literature. This has created a disparity in the quality of automatic evaluation methods for non-English languages, ultimately hindering the development of models with better multilingual capabilities. To bridge this gap, we introduce M-Prometheus, a suite of open-weight LLM judges ranging from 3B to 14B parameters that can provide both direct assessment and pairwise comparison feedback on multilingual outputs. M-Prometheus models outperform state-of-the-art open LLM judges on multilingual reward benchmarks spanning more than 20 languages, as well as on literary machine translation (MT) evaluation covering 4 language pairs. Furthermore, M-Prometheus models can be leveraged at decoding time to significantly improve generated outputs across all 3 tested languages, showcasing their utility for the development of better multilingual models. Lastly, through extensive ablations, we identify the key factors for obtaining an effective multilingual judge, including backbone model selection and training on synthetic multilingual feedback data instead of translated data. We release our models, training dataset, and code.

cs.CL↗

arXiv2025

XL-Suite: Cross-Lingual Synthetic Training and Evaluation Data for Open-Ended Generation

Cross-lingual open-ended generation - responding in a language different from that of the query - is an important yet understudied problem. This work proposes XL-Instruct, a novel technique for generating high-quality synthetic data, and introduces XL-AlpacaEval, a new benchmark for evaluating cross-lingual generation capabilities of large language models (LLMs). Our experiments show that fine-tuning with just 8K instructions generated using XL-Instruct significantly improves model performance, increasing the win rate against GPT-4o-Mini from 7.4% to 21.5% and improving on several fine-grained quality metrics. Moreover, base LLMs fine-tuned on XL-Instruct exhibit strong zero-shot improvements to question answering in the same language, as shown on our machine-translated m-AlpacaEval. These consistent gains highlight the promising role of XL-Instruct in the post-training of multilingual LLMs. Finally, we publicly release XL-Suite, a collection of training and evaluation data to facilitate research in cross-lingual open-ended generation.

cs.CL↗

arXiv2025

Tower+: Bridging Generality and Translation Specialization in Multilingual LLMs

Fine-tuning pretrained LLMs has been shown to be an effective strategy for reaching state-of-the-art performance on specific tasks like machine translation. However, this process of adaptation often implies sacrificing general-purpose capabilities, such as conversational reasoning and instruction-following, hampering the utility of the system in real-world applications that require a mixture of skills. In this paper, we introduce Tower+, a suite of models designed to deliver strong performance across both translation and multilingual general-purpose text capabilities. We achieve a Pareto frontier between translation specialization and multilingual general-purpose capabilities by introducing a novel training recipe that builds on Tower (Alves et al., 2024), comprising continued pretraining, supervised fine-tuning, preference optimization, and reinforcement learning with verifiable rewards. At each stage of training, we carefully generate and curate data to strengthen performance on translation as well as general-purpose tasks involving code generation, mathematics problem solving, and general instruction-following. We develop models at multiple scales: 2B, 9B, and 72B. Our smaller models often outperform larger general-purpose open-weight and proprietary LLMs (e.g., Llama 3.3 70B, GPT-4o). Our largest model delivers best-in-class translation performance for high-resource languages and top results in multilingual Arena Hard evaluations and in IF-MT, a benchmark we introduce for evaluating both translation and instruction-following. Our findings highlight that it is possible to rival frontier models in general capabilities, while optimizing for specific business domains, such as translation and localization.

cs.CL↗

arXiv2025

Adding Chocolate to Mint: Mitigating Metric Interference in Machine Translation

As automatic metrics become increasingly stronger and widely adopted, the risk of unintentionally "gaming the metric" during model development rises. This issue is caused by metric interference (MINT), i.e., the use of the same or related metrics for both model tuning and evaluation. MINT can misguide practitioners into being overoptimistic about the performance of their systems: as system outputs become a function of the interfering metric, their estimated quality loses correlation with human judgments. In this work, we analyze two common cases of MINT in machine translation-related tasks: filtering of training data, and decoding with quality signals. Importantly, we find that MINT strongly distorts instance-level metric scores, even when metrics are not directly optimized for-questioning the common strategy of leveraging a different, yet related metric for evaluation that is not used for tuning. To address this problem, we propose MINTADJUST, a method for more reliable evaluation under MINT. On the WMT24 MT shared task test set, MINTADJUST ranks translations and systems more accurately than state-of-the-art metrics across a majority of language pairs, especially for high-quality systems. Furthermore, MINTADJUST outperforms AUTORANK, the ensembling method used by the organizers.

cs.CL↗

arXiv2025

EuroLLM-9B: Technical Report

This report presents EuroLLM-9B, a large language model trained from scratch to support the needs of European citizens by covering all 24 official European Union languages and 11 additional languages. EuroLLM addresses the issue of European languages being underrepresented and underserved in existing open large language models. We provide a comprehensive overview of EuroLLM-9B's development, including tokenizer design, architectural specifications, data filtering, and training procedures. We describe the pre-training data collection and filtering pipeline, including the creation of EuroFilter, an AI-based multilingual filter, as well as the design of EuroBlocks-Synthetic, a novel synthetic dataset for post-training that enhances language coverage for European languages. Evaluation results demonstrate EuroLLM-9B's competitive performance on multilingual benchmarks and machine translation tasks, establishing it as the leading open European-made LLM of its size. To support open research and adoption, we release all major components of this work, including the base and instruction-tuned models, the EuroFilter classifier, and the synthetic post-training dataset.

cs.CL↗

arXiv2025

CroissantLLM: A Truly Bilingual French-English Language Model

We introduce CroissantLLM, a 1.3B language model pretrained on a set of 3T English and French tokens, to bring to the research and industrial community a high-performance, fully open-sourced bilingual model that runs swiftly on consumer-grade local hardware. To that end, we pioneer the approach of training an intrinsically bilingual model with a 1:1 English-to-French pretraining data ratio, a custom tokenizer, and bilingual finetuning datasets. We release the training dataset, notably containing a French split with manually curated, high-quality, and varied data sources. To assess performance outside of English, we craft a novel benchmark, FrenchBench, consisting of an array of classification and generation tasks, covering various orthogonal aspects of model performance in the French Language. Additionally, rooted in transparency and to foster further Large Language Model research, we release codebases, and dozens of checkpoints across various model sizes, training data distributions, and training steps, as well as fine-tuned Chat models, and strong translation models. We evaluate our model through the FMTI framework, and validate 81 % of the transparency criteria, far beyond the scores of even most open initiatives. This work enriches the NLP landscape, breaking away from previous English-centric work in order to strengthen our understanding of multilinguality in language models.

cs.CL↗

arXiv2025

EuroBERT: Scaling Multilingual Encoders for European Languages

General-purpose multilingual vector representations, used in retrieval, regression and classification, are traditionally obtained from bidirectional encoder models. Despite their wide applicability, encoders have been recently overshadowed by advances in generative decoder-only models. However, many innovations driving this progress are not inherently tied to decoders. In this paper, we revisit the development of multilingual encoders through the lens of these advances, and introduce EuroBERT, a family of multilingual encoders covering European and widely spoken global languages. Our models outperform existing alternatives across a diverse range of tasks, spanning multilingual capabilities, mathematics, and coding, and natively supporting sequences of up to 8,192 tokens. We also examine the design decisions behind EuroBERT, offering insights into our dataset composition and training pipeline. We publicly release the EuroBERT models, including intermediate training checkpoints, together with our training framework.

cs.CL↗

arXiv2025

WMT24++: Expanding the Language Coverage of WMT24 to 55 Languages & Dialects

As large language models (LLM) become more and more capable in languages other than English, it is important to collect benchmark datasets in order to evaluate their multilingual performance, including on tasks like machine translation (MT). In this work, we extend the WMT24 dataset to cover 55 languages by collecting new human-written references and post-edits for 46 new languages and dialects in addition to post-edits of the references in 8 out of 9 languages in the original WMT24 dataset. The dataset covers four domains: literary, news, social, and speech. We benchmark a variety of MT providers and LLMs on the collected dataset using automatic metrics and find that LLMs are the best-performing MT systems in all 55 languages. These results should be confirmed using a human-based evaluation, which we leave for future work.

cs.CL↗

arXiv2025

Translate Smart, not Hard: Cascaded Translation Systems with Quality-Aware Deferral

Larger models often outperform smaller ones but come with high computational costs. Cascading offers a potential solution. By default, it uses smaller models and defers only some instances to larger, more powerful models. However, designing effective deferral rules remains a challenge. In this paper, we propose a simple yet effective approach for machine translation, using existing quality estimation (QE) metrics as deferral rules. We show that QE-based deferral allows a cascaded system to match the performance of a larger model while invoking it for a small fraction (30% to 50%) of the examples, significantly reducing computational costs. We validate this approach through both automatic and human evaluation.

cs.CL↗