Malta

Despite the increasing demand for AI-based mental health monitoring tools, their practical utility for clinicians is limited by the lack of interpretability. The CLPsych 2024 Shared Task (Chim et al., 2024) aims to enhance the interpretability of Large Language Models (LLMs), particularly in mental health analysis, by providing evidence of suicidality through linguistic content. We propose a dual-prompting approach: (i) Knowledge-aware evidence extraction by leveraging the expert identity and a suicide dictionary with a mental health-specific LLM; and (ii) Evidence summarization by employing an LLM-based consistency evaluator. Comprehensive experiments demonstrate the effectiveness of combining domain-specific information, revealing performance improvements and the approach&#39;s potential to aid clinicians in assessing mental state progression.

EACL 2024

A Dual-Prompting for Interpretable Mental Health Language Models

large language model (llm)

mental health

interpretability

Truth and Responsibility
“Truth and Responsibility” is a call to reimagine anthropology to meet the demands of the present moment. The imperative to bear witness, take action, and be held accountable to the truths we write and circulate invites us to reflect on our responsibility in reckoning with disciplinary histories, harms, and possibilities. To whom are we giving evidence and toward what ends? For whom are we writing? To whom are we accountable, and in what ways? We are thrilled to announce the theme of the 2021 AAA Annual Meeting to be held in Baltimore, MD: Truth and Responsibility.

Many consider anthropology to be a holistic social science dedicated to cross-cultural understanding as a way to reduce bigotry and more fully comprehend “what it means to be human.” Others push against this readily circulated origin story, arguing that “human” remains a highly contested, political category—evident as the presumed set of rights and protections associated with the “human” continue to be stripped away violently in the communities in which some of us live and work. Which violent histories and epistemologies must anthropologists take responsibility for in telling truth(s) about the discipline’s development? How do the dual histories of settler colonialism and slavery continue to influence anthropological thought and practices? In these political times, how can anthropologists throughout the globe work to secure a capacious, progressive vision of the human? How might movements for anti-racism, decoloniality, queer liberation, and healing such as #BlackLivesMatter, #ProtectMaunaKea, and #MeToo push a future anthropology out of the ashes of the anthropological past? Finally, what are the limits and possibilities of the anthropological imagination?

Across subfields, we find truths in patterns of human behavior, language, evolution, and cultural worlds. Industry-positioned anthropologists do this by extending the boundaries of the discipline not as an administrative solution to job shortages but as a source of intervention and knowledge production. Increasingly, anthropologists who make use of the scientific method question how status and identity help to decide which facts come to represent truth. How might quantitative researchers use our methods to identify patterned truths, and what responsibility do we hold in challenging conventional wisdom about these patterns? What responsibility do we have to both our colleagues and larger publics, to push toward increased transparency and ethical grounding in our data collection, analysis, and presentation?

Pedagogy is a key medium for the communication of anthropological truths. Which (and whose) truths are foregrounded in our curricula? What are the possibilities for a liberatory pedagogy in a precarious, “post-truth” era? Conferences, museums, organizing spaces, and classrooms are sites where pedagogy and various forms of communication can help us interrogate the performativity of truth. How might anthropologists conceive of truth performances? Might visual, aural, and choreographic performances during AAA sessions offer anthropologists alternative genres for truth-telling and truth-making that provoke discussions about responsibility?

In accordance with the urgency of the year’s theme, the questions posed in this call for papers do not have easy answers. The hope is that those who submit proposals will take up these questions and engage in collective thinking and imagining about the truths we hold, the truths we challenge, and the responsibilities we bear in co-creating a free and more equitable world. All panel proposals should have a statement about how the panel has incorporated the goals of equity, diversity, and inclusion, and/or analysis of power.

Executive Session proposals have already been selected and speak directly to the theme “Truth and Responsibility.” To practice the critical reflexivity this call for papers invites, we asked that Executive Session submissions explicitly address the diversity of session participants, such as organizational/institutional affiliation, career stage, race, ethnicity, disability, sexuality, class, religion, national identity, and gender. Proposals for Executive Sessions that demonstrated a commitment to equity and inclusion and/or a clear analysis of power (i.e. anti-imperialist, anti-white supremacist, decolonial, anti-ableist, feminist, anti-transphobic, anti-xenophobic, etc) were given priority. We are excited to be in conversation with residents of the Baltimore region and look forward to the discussions, performances, and actions generated by the dynamic group of anthropologists attending the 2021 AAA Annual Meeting.

AAA Annual Meeting 2021

“Truth and Responsibility” is a call to reimagine anthropology to meet the demands of the present moment. The imperative to bear witness, take action, and be held accountable to the truths we write and circulate invites us to reflect on our responsibility in reckoning with disciplinary histories, harms, and possibilities. To whom are we giving evidence and toward what ends? For whom are we writing? To whom are we accountable, and in what ways? We are thrilled to announce the theme of the 2021 AAA Annual Meeting to be held in Baltimore, MD: Truth and Responsibility.

Despite the increasing demand for AI-based mental health monitoring tools, their practical utility for clinicians is limited by the lack of interpretability. The CLPsych 2024 Shared Task (Chim et al., 2024) aims to enhance the interpretability of Large Language Models (LLMs), particularly in mental health analysis, by providing evidence of suicidality through linguistic content. We propose a dual-prompting approach: (i) Knowledge-aware evidence extraction by leveraging the expert identity and a suicide dictionary with a mental health-specific LLM; and (ii) Evidence summarization by employing an LLM-based consistency evaluator. Comprehensive experiments demonstrate the effectiveness of combining domain-specific information, revealing performance improvements and the approach's potential to aid clinicians in assessing mental state progression.

workshop paper

Welcome to the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL). Continuing its mission of expanding and involving the science community of all European countries, EACL has selected the Malta community for the 18th EACL. Considering the importance of physical interaction among researchers, the conference will be held at the Hotel Radisson Blu, St. Julians, in Malta, from 17 to 22 of March, 2024. The conference will also feature the possibility to participate virtually. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language. 
#### [Conference Handbook](https://drive.google.com/file/d/13Wn38q0yev6U3RTgnf4Tn5T9GuAuZ-xW/view?usp=sharing) *downloadable*

To access the EACL 2024 event page you need to register [**here**](https://acl.swoogo.com/EACL2024). 

Please register!

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

Hate speech refers to the offensive remarks against a community or individual based on inherent characteristics. Hate speech against a community based on their caste and native are unfortunately prevalent in the society. Especially with social media platforms being a very popular tool for communication and sharing ideas, people post hate speech against caste or immigrants on social medias. The Shared Task LT–EDI 2024: Caste/Immigration Hate Speech Detection\cite{chakravarthi-caste-2024-overview} was created with the objective to create an automatic classification system that detects and classifies hate speech posted on social media targeting a community belonging to a particular caste and immigrants. Datasets in Tamil language were provided along with the shared task. We experimented with several traditional models such as Naive Bayes, Support Vector Machine (SVM), Logistic Regression, Random Forest Classifier and Decision Tree Classifier out of which Support Vector Machine yielded the best results placing us 8th in the rank list released by the organizers.

Quartet@LT-EDI 2024: A Support Vector Machine Approach For Caste/Immigration Hate Speech Detection

This paper presents our submission for Shared task on Stress Identification in Dravidian Languages: StressIdent LT-EDI@EACL2024. The objective of this task is to identify stress levels in individuals based on their social media content. The system is tasked with analysing posts written in a code-mixed language of Tamil and Telugu and categorising them into two labels: "stressed" or "not stressed." Our approach aimed to leverage feature extraction and juxtapose the performance of widely used traditional, deep learning and transformer models. Our research highlighted that building a pipeline with traditional classifiers proved to significantly improve their performance (0.98 and 0.93 F1-scores in Telugu and Tamil respectively), surpassing the baseline as well as deep learning and transformer models.

SSN-Nova@LT-EDI 2024: Leveraging Vectorisation Techniques in an Ensemble Approach for Stress Identification in Low-Resource Languages

Meme is a very popular term prevailing among almost all social media platforms in recent days. A meme can be a combination of text and image whose sole purpose is meant to be funny and entertain people. Memes can sometimes promote misogynistic content expressing hatred, contempt or prejudice against women. The Shared Task LT–EDI 2024: Multitask Meme classification – Unraveling Misogynistic and Trolls in Online Memes Task 1 was created with the purpose to classify social media memes as "Misogynistic" and "Non - Misogynistic". The task encompassed Tamil and Malayalam datasets. We separately classified the textual data using Multinomial Naive Bayes and pictorial data using ResNet50 model. The results of from both data were combined to yield an overall result. We were ranked 2nd for both languages in this task.

Quartet@LT-EDI 2024: A SVM-ResNet50 Approach For Multitask Meme classification - Unraveling Misogynistic and Trolls in Online Memes

Homophobia and transphobia are terms which are used to describe the fear or hatred towards people who are attracted to the same sex or people whose psychological gender differs from his biological sex. People use social media to exert this behaviour. The increased amount of abusive content negatively affects people in a lot of ways. It makes the environment toxic and unpleasant to LGBTQ+ people. The paper talks about the classification model for classifying the contents into 3 categories which are homophobic, transphobic and non-homophobic/transphobic. We used many traditional models like Support Vector Machine, Random Classifier, Logistic Regression and K- Nearest Neighbour to achieve this. The macro average F1 scores for Malayalam, Telugu, English, Marathi, Kannada, Tamil, Gujarati, Hindi are 0.88, 0.94, 0.96, 0.78, 0.93, 0.77, 0.94, 0.47 and the rank for these languages are 5, 6, 9, 6, 8, 6, 6, 4.

Quartet@LT-EDI 2024: Support Vector Machine based approach for Homophobia/Transphobia Detection in social media comments

This paper presents our submission for the shared task on Caste/Immigration Hate Speech Detection: LT-EDI@EACL 2024. This text classification task aims to foster the creation of models capable of identifying hate speech related to caste and immigration. The dataset comprises of social media comments, and the goal is to categorise them into negative and positive sentiments. Our approach explores back-translation for data augmentation to address sparse datasets in low-resource Dravidian languages. While Part-of-Speech (POS) tagging is valuable in natural language processing, our work highlights its ineffectiveness in Dravidian languages, with model performance drastically reducing from 0.73 to 0.67 on application. In analysing boosting and ensemble methods, the voting classifier with traditional models outperforms others and the boosting techniques, underscoring the efficacy of simpler models on low-resource data despite augmentation.

SSN-Nova@LT-EDI 2024: POS Tagging, Boosting Techniques and Voting Classifiers for Caste/Immigration Hate Speech Detection

The widespread use of online communication has caused a significant increase in the spread of hate speech on social media. However, there are also hate crimes based on caste or immigration status. Despite India's efforts to bring equality among its citizens, numerous crimes occur just based on caste. Immigration-based hostility happens both in India and in developed countries. A shared task was arranged to address this issue in a low-resource language such as Tamil. This paper aims to improve the detection of hate speech and hostility based on caste and immigration status on social media. To achieve this, we investigated several Machine Learning (ML), Deep Learning (DL), and transformer-based models, including M-BERT, XLM-R, and Tamil BERT. We obtained a macro -score of 0.80 using the M-BERT model, which enabled us to rank on the shared task.

CUET_NLP_Manning_LT-EDI@EACL2024: Transformer Based Approach on Caste or Immigration Hate Speech Detection

This paper describes the Unibuc Archaeology team work for CLPsych's 2024 Shared Task that involved finding evidence within the text supporting the assigned suicide risk level. Two types of evidence were required: highlights (extracting relevant spans within the text) and summaries (aggregating evidence into a synthesis). Our work focuses on evaluating Large Language Models (LLM) as opposed to an alternative method that is much more memory and resource efficient. The first approach employs an LLM that is used for generating the summaries and is guided to provide sequences of text indicating suicidal tendencies through a processing chain for highlights. The second approach involves implementing a good old-fashioned machine learning tf-idf with a logistic regression classifier, whose representative features we use to extract relevant highlights.

Cheap Ways of Extracting Clinical Markers from Texts

This paper presents our contribution to the CLPsych 2024 shared task, focusing on the use of open-source large language models (LLMs) for suicide risk assessment through the analysis of social media posts. We achieved first place (out of 15 participating teams) in the task of providing summarized evidence of a user's suicide risk. Our approach is based on Retrieval Augmented Generation (RAG), where we retrieve the top-k (k=5) posts with the highest emotional charge and provide the level of three different negative emotions (sadness, fear, anger) for each post during the generation phase.

Utilizing Large Language Models to Identify Evidence of Suicidality Risk through Analysis of Emotionally Charged Posts

We propose a method that integrates supervised extractive and generative language models for providing supporting evidence of suicide risk in the CLPsych 2024 shared task. Our approach comprises three steps. Initially, we construct a BERT-based model for estimating sentence-level suicide risk and negative sentiment. Next, we precisely identify high suicide risk sentences by emphasizing elevated probabilities of both suicide risk and negative sentiment. Finally, we integrate generative summaries using the MentaLLaMa framework and extractive summaries from identified high suicide risk sentences and a specialized dictionary of suicidal risk words. SophiaADS, our team, achieved 1st place for highlight extraction and ranked 10th for summary generation, both based on recall and consistency metrics, respectively.

Integrating Supervised Extractive and Generative Language Models for Suicide Risk Evidence Summarization

Research on psychological risk factors for suicide has developed for decades. However, combining explainable theory with modern data-driven language model approaches is non-trivial. In this study, we propose and evaluate methods for identifying language patterns aligned with theories of suicide risk by combining theory-driven suicidal archetypes with language model-based and relative entropy-based approaches. Archetypes are based on prototypical statements that evince risk of suicidality while relative entropy considers the ratio of how unusual both a risk-familiar and unfamiliar model find the statements. While both approaches independently performed similarly, we find that combining the two significantly improved the performance in the shared task evaluations, yielding our combined system submission with a BERTScore Recall of 0.906. Consistent with the literature, we find that titles are highly informative as suicide risk evidence, despite the brevity. We conclude that a combination of theory- and data-driven methods are needed in the mental health space and can outperform more modern prompt-based methods.

Downloads

Next from EACL 2024

Quartet@LT-EDI 2024: A Support Vector Machine Approach For Caste/Immigration Hate Speech Detection

Similar lecture

New and Emerging Perspectives in the Anthropology of Mental Health II

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EACL 2024

Quartet@LT-EDI 2024: A Support Vector Machine Approach For Caste/Immigration Hate Speech Detection

Similar lecture

New and Emerging Perspectives in the Anthropology of Mental Health II

Downloads