The integration of large language models (LLMs) into mental health applications offers promising opportunities for positive social impact. However, it also presents critical risks. While previous studies have often addressed these challenges and risks individually, a broader and multi-dimensional approach is still lacking. In this paper, we introduce a taxonomy of the main challenges related to the use of LLMs for mental health and propose a structured, comprehensive research agenda to mitigate them. We emphasize the need for explainable, emotionally aware, culturally sensitive, and clinically aligned systems, supported by continuous monitoring and human oversight. By placing our work within the broader context of natural language processing (NLP) for positive impact, this research contributes to ongoing efforts to ensure that technological advances in NLP responsibly serve vulnerable populations, fostering a future where mental health solutions improve rather than endanger well-being.

Large language models (LLMs) often inherit and amplify social biases embedded in their training data. A prominent social bias is gender bias. In this regard, prior work has mainly focused on gender stereotyping bias – the association of specific roles or traits with a particular gender – in English and on evaluating gender bias in model embeddings or generated outputs. In contrast, gender representation bias – the unequal frequency of references to individuals of different genders – in the training corpora has received less attention. Yet such imbalances in the training data constitute an upstream source of bias that can propagate and intensify throughout the entire model lifecycle. To fill this gap, we propose a novel LLM-based method to detect and quantify gender representation bias in LLM training data in gendered languages, where grammatical gender challenges the applicability of methods developed for English. By leveraging the LLMs’ contextual understanding, our approach automatically identifies and classifies person-referencing words in gendered language corpora. Applied to four Spanish-English benchmarks and five Valencian corpora, our method reveals substantial male-dominant imbalances. We show that such biases in training data affect model outputs, but can surprisingly be mitigated leveraging small-scale training on datasets that are biased towards the opposite gender. Our findings highlight the need for corpus-level gender bias analysis in multilingual NLP. We make our code and data publicly available.

The widespread adoption of chat interfaces based on Large Language Models (LLMs) raises concerns about promoting superficial learning and undermining the development of critical thinking skills. Instead of relying on LLMs purely for retrieving factual information, this work explores their potential to foster deeper reasoning by generating critical questions that challenge unsupported or vague claims in debate interventions. This study is part of a shared task of the 12th Workshop on Argument Mining, co-located with ACL 2025, focused on automatic critical question generation. We propose a two-step framework involving two small-scale open source language models: a Questioner that generates multiple candidate questions and a Judge that selects the most relevant ones. Our system ranked first in the shared task competition, demonstrating the potential of the proposed LLM-based approach to encourage critical engagement with argumentative texts.