Indonesia

Identifying mental health conditions using user-generated text on social media paved the way for automated computational methods for mental health surveillance on social media. Inferring the accurate state of mental health markers requires understanding the user&#39;s emotions and history of mental health conditions for identifying the mental health landscape of at-risk users. Existing works have exploited social media data by using users&#39; most recent posts; however, previous works ignored users&#39; emotional historical activity on social media, indicating their mental state over time. In this work, we address this gap and present a novel emotion- and time-aware architecture that jointly learns social media users&#39; emotional historical context and temporal posting irregularities for mental health surveillance on social media. We conduct classification experiments to identify social media users with mental conditions, i.e., (i) healthy users and users affected with depression and (ii) healthy users and users prone to self-harm. Experimental results demonstrate proposed method outperforms recent competitive methods, demonstrating the importance of capturing the user&#39;s temporal emotional spectrum and time-aware emotional context through historical posts for social media users&#39; mental health surveillance.


IJCNLP-AACL 2023

Temporal Tides of Emotional Resonance: A Novel Approach to Identify Mental Health on Social Media

### Welcome to CogSci Conference 2025!

The 47th Annual Meeting of the Cognitive Science Society was a hybrid meeting held in San Francisco. 

<div style="position:relative;padding-top:0;width:900px;height:500px;"><iframe style="position:absolute;border:none;width:100%;height:100%;left:0;top:0;" src="https://online.fliphtml5.com/ebtyf/amvr/" seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe></div>

#### About

The Cognitive Science Society brings together researchers from around the world who hold a common goal: understanding the nature of the human mind. The mission of the Society is to promote Cognitive Science as a discipline, and to foster scientific interchange among researchers in various areas of study, including Artificial Intelligence, Linguistics, Anthropology, Psychology, Neuroscience, Philosophy, and Education.

The Society is a non-profit professional organization and its activities include sponsoring an annual conference and publishing the journals Cognitive Science and TopiCS.

#### Our History 

* **Society Creation** 
The Society was incorporated as a 501(c)(3) non-profit professional organization in Massachusetts in 1979. The organizing committee included Roger Schank, Allan Collins, Donald Norman, and a number of other scholars from psychology, linguistics, computer science, and philosophy. 
 
* **Conference Creation** 
The first conference on cognitive science was held at La Jolla, California in August, 1979, and has occurred annually since then. The proceedings of each conference are published, and those from most years are available through Lawrence Erlbaum Associates, Inc. The annual proceedings of the Cognitive Science Conference represent a major source of information on new work and new ideas in the scientific study of thinking. In 1990, the Society, with help from an anonymous donor, established the David Marr Prize for the best student paper at each annual meeting.
 
* **Journal Creation** 
The Journal, Cognitive Science, began publication in 1976, and is now published by Wiley-Blackwell. The Executive Editor is currently Richard P. Cooper of Birkbeck, University of London, and there are 18 Associate Editors and a 30-member editorial board. It serves as the premier outlet for research reports that intersect two or more disciplines. Copyrights for articles published in the journal are held by the Society. The Governing Board of the Cognitive Science Society voted in late 2006 to found a new journal, Topics in Cognitive Science (topiCS). The Editor in Chief is Wayne Gray, Cognitive Science Department, Rensselaer Polytechnic Institute. The journal seeks to fill a niche not occupied by Cognitive Science Journal or other cognitive science journals. Membership in the Society includes a subscription to Cognitive Science and TopiCS. Copyrights for articles published in the journal are held by the Society.
 

#### Code of Conduct

By attending the CogSci 2025 Conference, you are required to adhere to the society’s **[Code of Conduct](https://drive.google.com/file/d/1ChPuihLy6jE_BWqfO7J2KKgX35JW2zsM/view?usp=sharing)**.

You need to log in with the email address you registered with. 

Login credentials were sent to you from Underline -  subject line "Welcome to the CogSci 2025 Conference". Please be sure to check your spam/promotional inbox  if you do not see an email confirmation right away.





Please log in to join this event.

To access the site, please register [**here**](https://cognitivesciencesociety.org/registration/).

If you are registered and feel like you are seeing this message by mistake, please make sure you are logged in with the same email that you registered with. 

Please register!

CogSci 2025

The 47th Annual Meeting of the Cognitive Science Society presents the latest research across cognitive science and highlights the theme of Cognition in Context.

Identifying mental health conditions using user-generated text on social media paved the way for automated computational methods for mental health surveillance on social media. Inferring the accurate state of mental health markers requires understanding the user's emotions and history of mental health conditions for identifying the mental health landscape of at-risk users. Existing works have exploited social media data by using users' most recent posts; however, previous works ignored users' emotional historical activity on social media, indicating their mental state over time. In this work, we address this gap and present a novel emotion- and time-aware architecture that jointly learns social media users' emotional historical context and temporal posting irregularities for mental health surveillance on social media. We conduct classification experiments to identify social media users with mental conditions, i.e., (i) healthy users and users affected with depression and (ii) healthy users and users prone to self-harm. Experimental results demonstrate proposed method outperforms recent competitive methods, demonstrating the importance of capturing the user's temporal emotional spectrum and time-aware emotional context through historical posts for social media users' mental health surveillance.


workshop paper

### Welcome to IJCNLP-AACL 2023!
I am honored and happy to welcome you to IJCNLP-AACL 2023! The conference hosts both the 13th
International Joint Conference on Natural Language Processing (IJCNLP) and the 3rd Conference of
the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL). 
The joint conferences of IJCNLP and AACL have been organized by alternating leadership in the
Asia-Pacific region, in odd years by Asian Federation of Natural Language Processing (AFNLP) and
in even years by AACL. The conference is organized by AFNLP this year. 

Considering the current situation that we are not yet fully over the coronavirus pandemic, the
conference is offered in a hybrid mode, off-line in Nusa Dua, Bali, Indonesia, and on-line with the help
of Underline, during November 1-4, 2023. For those attending off-line, I hope that you will find time to
explore the exciting local area, which remains quite warm and pleasant at this time of the year. For
those attending on-line, I hope that the attendance is not much hampered, sometimes even enhanced
and comfortable, though the on-line arrangements have been made without ample time. 

Message from Jong C. Park, Korea Advanced Institute of Science and Technology (KAIST) 
IJCNLP-AACL 2023 General Chair 
###### [Conference Handbook](https://drive.google.com/file/d/1S2JSXa105xJxIkNBfhYmGH20Om6wARBx/view?usp=sharing)

In order to acces this event page, you need to register [here](http://www.ijcnlp-aacl2023.org/)

The 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics

Controllable text generation is a fundamental aspect of natural language generation, with numerous methods proposed for different constraint types. However, these approaches often require significant architectural or decoding modifications, making them challenging to apply to additional constraints or resolve different constraint combinations. To address this, our paper introduces Regular Expression Instruction (REI), which utilizes an instruction-based mechanism to fully exploit regular expressions' advantages to uniformly model diverse constraints. Specifically, our REI supports all popular fine-grained controllable generation constraints, i.e., lexical, positional, and length, as well as their complex combinations, via regular expression-style instructions. Our method only requires fine-tuning on medium-scale language models or few-shot, in-context learning on large language models, and requires no further adjustment when applied to various constraint combinations. Experiments demonstrate that our straightforward approach yields high success rates and adaptability to various constraints while maintaining competitiveness in automatic metrics and outperforming most previous baselines.

Toward Unified Controllable Text Generation via Regular Expression Instruction

Solving math word problems (MWP) is a challenging task for natural language processing systems, as it requires to not only identify and comprehend the problem description within the context, but also to deduce a solution in accordance with the posed question. Previous solvers have been found to prioritize the context over the question, resulting in low performance when solving multiple questions under the same context. In this paper, we present a question-oriented strategy to address this issue and improve the generalizability of MWP solvers. Our approach features an entity-aware encoder that enhances the connection between MWP context and question via entities in established dependency graphs, aiming at obtaining better problem representations. Then, a question-guided decoder is trained using a contrastive learning strategy to enhance the question representations. Empirical evaluations on four benchmarks demonstrate that our method outperforms previous solvers and exhibits a favorable balance between efficacy and efficiency in MWP solving. In addition, our solver is not reliant on any specific pre-trained model and demonstrates seamless compatibility with different pre-trained model backbones.

Don't be Blind to Questions: Question-Oriented Math Word Problem Solving

Detecting out-of-scope (OOS) utterances is crucial in task-oriented dialogue systems, but obtaining enough annotated OOS dialogues to train a binary classifier directly is difficult in practice. Existing data augmentation methods generate OOS dialogues automatically, but their performance usually depends on an external corpus. This dependence not only induces uncertainty, but also reduces the quality of generated dialogues. Specifically, all of them are out-of-domain (OOD). 

Herein we propose SILVER, a self data augmentation method that does not use external data. It addresses issues of previous research and improves the accuracy of OOS detection (false positive rate: 90.5% to 47.4%). Furthermore, SILVER successfully generates high-quality in-domain (IND) OOS dialogues in terms of naturalness (percentage: 8% to 68%) and OOS correctness (percentage: 74% to 88%), as evaluated by human workers.

SILVER: Self Data Augmentation for Out-of-Scope Detection in Dialogues

State-of-the-art summarization systems can generate highly fluent summaries. These summaries, however, may contain factual inconsistencies and/or information not present in the source. Hence, an important component of assessing the quality of summaries is to determine whether there is information consistency between the source and the summary. Existing approaches are typically based on lexical matching or representation-based methods. In this work, we introduce an alternative scheme based on standard information-theoretic measures in which the information present in the source and summary is directly compared. We propose a Multiple-choice Question Answering and Generation framework, MQAG, which approximates the information consistency by computing the expected statistical distance between summary and source answer distributions over automatically generated multiple-choice questions. This approach exploits multiple-choice answer probabilities, as predicted answer distributions can be compared. We conduct experiments on four summary evaluation datasets: QAG-CNNDM/XSum, XSum-Hallucination, Podcast Assessment, and SummEval. Experiments show that MQAG, using models trained on SQuAD or RACE, outperforms existing evaluation methods on the majority of tasks.

MQAG: Multiple-choice Question Answering and Generation for Assessing Information Consistency in Summarization

Generating high-quality non-English language datasets is crucial for achieving high performance in various Natural Language Processing (NLP) tasks. In this paper, we propose a new approach for translating NLP datasets that relies on a two-phase pipeline and online translation services. Our approach focuses on solving the alignment problem that affects span prediction tasks and utilizes automatically labeled data for training an alignment model. We demonstrate that our model-based approach shows higher accuracy than any other alignment method and improves the average F1 score on several Question-Answering (QA) datasets, specifically on the XQuAD Translated-train dataset, achieving new state-of-the-art results.

Automatic Translation of Span-Prediction Datasets

While both text and Knowledge Graphs (KG) may be used to answer a question, most current Question Answering and Generation models only work on a single modality. In this paper, we introduce a multi-task model such that questions can be generated and answered from both KG and text. The model has wide coverage and handles both simple (one KG fact) and complex (more than one KG fact) questions. Extensive internal, cross-modal and external consistency checks, and analysis of the quality of the generated questions, show that our approach outperforms previous work. Our data and modeling also leads to improvements in downstream tasks, including better performance with finetuning Open-Domain QA architectures and better correlation with human judgments than the Data-QuestEval metric which was previously proposed for evaluating the semantic adequacy of KG-to-Text generations.

Generating and Answering Simple and Complex Questions from Text and from Knowledge Graphs

The strong capabilities of conversation-based large language models in healthcare applications are attracting an increasingly larger audience. However, the reliability of these models in consistently and accurately providing medical advice based on user-inputted symptoms is a critical concern. This study explores the sensitivity of LLMs to variations in user input, focusing specifically on how different symptom descriptions and prior users' beliefs can potentially lead to different diagnoses. We test two GPT models with five different prompt templates to assess their ability of mentioning the true patient condition based on the symptoms. Our findings reveal a substantial sensitivity to input variations---especially when users have prior assumptions and beliefs---indicating potential inconsistencies in the diagnoses generated by these models.

“Dr LLM, what do I have?”: The Impact of User Beliefs and Prompt Formulation on Health Diagnoses

In clinical visits, clinical note writing is a time-consuming and cost-prohibitive manual task for clinicians. Although virtual medical scribes have been proposed to generate clinical notes (semi-)automatically, the data sparsity issue is still a challenging problem in practice. Identifying the topic of clinical utterances in doctor-patient conversations is one of the key strategies for automation. In this paper, we propose an utterance-level note section classification method for the situation of the limited amount of in-house data. We leverage an external, unsupervised corpus of medical conversations to transfer knowledge using the framework of Unsupervised Meta-learning with Task Augmentation (UMTA). Our experiments are performed on both manual transcripts and machine transcripts generated by automatic speech recognition (ASR). The results show that our strategies achieve substantial gains in prediction accuracy over several baseline approaches and are robust to ASR errors.

Clinical note section classification on doctor-patient conversations in low-resourced settings

Huge amounts of sentiment-rich data are produced on social networks, the rate of which is growing year by year. As such, a global view of sentiment on social media platforms is only possible at scale through automation. The huge data sizes also mean that sentiment analysis systems must balance both accuracy and computing resource requirements. Typical NLP research maximizes accuracy, while other metrics are overlooked and thus older models are forgotten. Prior models are easily forgotten despite their possible suitability in settings where large computing resources are unavailable or relatively more costly. In this paper, we perform a broad comparative evaluation of document-level sentiment analysis models with a focus on resource costs that are important for the feasibility of model deployment and general climate consciousness. Our experiments consider different feature extraction techniques, the effect of ensembling, task-specific deep learning modeling, and domain-independent large language models (LLMs). We find that while a fine-tuned LLM achieves the best accuracy, some alternate configurations provide huge (up to 24, 283×) resource savings for a marginal (<1%) loss in accuracy. Furthermore, we find that for smaller datasets, the differences in accuracy shrink while the difference in resource consumption grows further.

Efficient Sentiment Analysis: A Resource-Aware Evaluation of Feature Extraction Techniques, Ensembling, and Deep Learning Models

Code-mixing is a common linguistic phenomenon when words or syllables from two or more languages are used within a single sentence. Several works have been conducted on building such datasets and performing downstream tasks on them. Although it is very common to observe code-mixing with multiple languages, most datasets in this domain contain code-mixed data from only two languages. In this paper, we introduce OffMix-3L, a novel dataset for the task of offensive language identification which contains code-mixed data from 3 different languages. We also provide a comprehensive experimental analysis with several models and observe that BanglishBERT outperforms all the transformer-based models and also the generative model GPT-3.5 on OffMix-3L.

Downloads

Next from IJCNLP-AACL 2023

Toward Unified Controllable Text Generation via Regular Expression Instruction

Similar lecture

Integrating Textual and Emotional Dynamics for Accurate Detection of Mental Health Disorders in Social Media

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.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 IJCNLP-AACL 2023

Toward Unified Controllable Text Generation via Regular Expression Instruction

Similar lecture

Integrating Textual and Emotional Dynamics for Accurate Detection of Mental Health Disorders in Social Media

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads