Austria

Green industrial policies (GIPs) are government interventions that support environmentally sustainable economic growth through targeted incentives, regulations, and investments in clean technologies. As the backbone of climate mitigation and adaptation, GIPs deserve systematic documentation and analysis. However, two major hurdles impede this systematic documentation. First, unlike other climate policy documents, such as Nationally Determined Contributions (NDCs) which are centrally curated, GIPs are scattered across numerous government legislation and policy announcements. Second, extracting information from these diverse documents is expensive when relying on expert annotation. We address this gap by proposing \textit{GreenSpyder}, an LLM-based workflow that actively monitors, classifies, and annotates GIPs from open-source information. As a demonstration, we benchmark LLM performance in classifying and annotating GIPs on a small expert-curated dataset. Our results show that LLMs can be quite effective for classification and coarse annotation tasks, though they still need improvement for more nuanced classification. Finally, as a real-world application, we apply \textit{GreenSpyder} to U.S. Legislative Records from the 117th Congress, paving the way for more comprehensive LLM-based GIP documentation in the future.

ACL 2025

Tracking Green Industrial Policies with LLMs: A Demonstration

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

workshop paper

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Transformer-based Job Ranking Systems (JRSs) are vulnerable to societal biases inherited in unbalanced datasets. These biases often manifest as unjust job rankings, particularly disadvantaging candidates of different genders. Most bias mitigation techniques leverage candidates’ gender and align gender distributions within the embeddings of JRSs to mitigate bias. While such methods effectively align distributional properties and make JRSs agnostic to gender, they frequently fall short in addressing empirical fairness metrics, such as the performance gap across genders. In this study, we shift our attention from candidate gender to mitigate bias based on gendered language in job advertisements. We propose a novel neutrality score based on automatically discovered biased words in job ads and use it to re-rank the model’s decisions. We evaluate our method by comparing it with different bias mitigation strategies and empirically demonstrate that our proposed method not only improves fairness but can also enhance the model’s performance.


Mitigating Gender Bias in Job Ranking Systems Using Job Advertisement Neutrality

Advanced AI models that generate realistic images from text prompts offer new creative possibilities but also risk producing culturally insensitive or offensive content. To address this issue, we introduce a novel dataset designed to classify text prompts that could lead to the generation of harmful images misrepresenting different cultures and communities. By training machine learning models on this dataset, we aim to automatically identify and filter out harmful prompts before image generation, balancing cultural sensitivity with creative freedom. Benchmarking with state-ofthe-art language models, our baseline models achieved an accuracy of 73.34%.


AI Tools Can Generate Misculture Visuals! Detecting Prompts Generating Misculture Visuals For Prevention

Although the responses to events such as COVID-19 have been extensively studied, research on sudden crisis response in a multicultural context is still limited. In this paper, our contributions are 1)We examine cultural differences in social media posts related to such events in two different countries, specifically the United Kingdom lockdown of 2020-03-23 and the China Urumqi fire1 of 2022-11-24. 2) We extract the emotional polarity of tweets and weibos gathered temporally adjacent to those two events, by fine-tuning transformer-based language models for each language. We evaluate each model’s performance on 2 benchmarks, and show that, despite being trained on a relatively small amount of data, they exceed baseline accuracies. We find that in both events, the increase in negative responses is both dramatic and persistent, and does not return to baseline even after two weeks. Nevertheless, the Chinese dataset reflects, at the same time, positive responses to subsequent government action. Our study is one of the first to show how sudden crisis events can be used to explore affective reactions across cultures


Cross-cultural Sentiment Analysis of Social Media Responses to a Sudden Crisis Event

When a crisis hits, people often turn to social media to ask for help, offer help, find out how others are doing, and decide what they should do. The growth of social media use during crises has been helpful to aid providers as well, giving them a nearly immediate read of the on-the-ground situation that they might not otherwise have. The amount of crisis-related content posted to social media over the past two decades has been explosive, which, in turn, has been a boon to Language Technology (LT) researchers. In this study, we conducted a systematic survey of 355 papers published in the past five years to better understand the expanding growth of LT as it is applied to crisis content, specifically focusing on corpora built over crisis social media data as well as systems and applications that have been developed on this content. We highlight the challenges and possible future directions of research in this space. Our goal is to engender interest in the LT field writ large, in particular in an area of study that can have dramatic impacts on people’s lives. Indeed, the use of LT in crisis response has already been shown to save people’s lives.


Tapping into Social Media in Crisis: A Survey

Effective emotional support in conversation requires strategic decision making, as it involves complex, context-sensitive reasoning tailored to diverse individual needs. The Emotional Support Conversation framework addresses this by organizing interactions into three distinct phases—exploration, comforting, and action—which guide strategy selection during response generation. While multitask learning has been applied to jointly optimize strategy prediction and response generation, it often suffers from task interference due to conflicting learning objectives. To overcome this, we propose the Strategy-Aware Refinement Module (STAR), which disentangles the decoder’s hidden states for each task and selectively fuses them via a dynamic gating mechanism. This design preserves task-specific representations while allowing controlled information exchange between tasks, thus reducing interference. Experimental results demonstrate that STAR effectively reduces task interference and achieves state-of-the-art performance in both strategy prediction and supportive response generation.


STAR: Strategy-Aware Refinement Module in Multitask Learning for Emotional Support Conversations

Large Language Models (LLMs) have shown promise in automating high-labor data tasks, but the adoption of LLMs in high-stake scenarios faces two key challenges: their tendency to answer despite uncertainty and their difficulty handling long input contexts robustly. We investigate commonly used off-the-shelf LLMs’ ability to identify low-confidence outputs for human review through check set selection–a process where LLMs prioritize information needing human judgment. Using a case study on social media monitoring for disaster risk management, we define the “check set” as a list of tweets escalated to the disaster manager when the LLM has the least confidence, enabling human oversight within budgeted effort. We test two strategies for LLM check set selection: individual confidence elicitation – LLM assesses confidence for each tweet classification individually, requiring more prompts with shorter contexts, and direct set confidence elicitation – LLM evaluates confidence for a list of tweet classifications at once, using less prompts but longer contexts. Our results reveal that set selection via individual probabilities is more reliable but that direct set confidence merits further investigation. Direct set selection challenges include inconsistent outputs, incorrect check set size, and low inter-annotator agreement. Despite these challenges, our approach improves collaborative disaster tweet classification by outperforming random-sample check set selection, demonstrating the potential of human-LLM collaboration.


Evaluating Large Language Models for Confidence-based Check Set Selection

The rise of LLM-driven AI characters raises safety concerns, particularly for vulnerable human users with psychological disorders. To address these risks, we propose EmoAgent, a multi-agent AI framework designed to evaluate and mitigate mental health hazards in humanAI interactions. EmoAgent comprises two components: EmoEval simulates virtual users, including those portraying mentally vulnerable individuals, to assess mental health changes before and after interactions with AI characters. It uses clinically proven psychological and psychiatric assessment tools (PHQ-9, PDI, PANSS) to evaluate mental risks induced by LLM. EmoGuard serves as an intermediary, monitoring users’ mental status, predicting potential harm, and providing corrective feedback to mitigate risks. Experiments conducted in popular character-based chatbots show that emotionally engaging dialogues can lead to psychological deterioration in vulnerable users, with mental state deterioration in more than 34.4% of the simulations. EmoGuard significantly reduces these deterioration rates, underscoring its role in ensuring safer AI-human interactions.


EmoAgent: Assessing and Safeguarding Human-AI Interaction for Mental Health Safety

The development of chatbots requires collecting a large number of human-chatbot dialogues to reflect the breadth of users' sociodemographic backgrounds and conversational goals. However, the resource requirements to conduct the respective user studies can be prohibitively high and often only allow for a narrow analysis of specific dialogue goals and participant demographics. In this paper, we propose LLM Roleplay, the first comprehensive method integrating multi-turn human-chatbot interaction simulation, explicit persona construction from sociodemographic traits, goal-driven dialogue planning, and robust handling of conversational failures, enabling broad utility and reliable dialogue generation. To validate our method, we collect natural human-chatbot dialogues from different sociodemographic groups and conduct a user study to compare these with our generated dialogues. We evaluate the capabilities of state-of-the-art LLMs in maintaining a conversation during their embodiment of a specific persona and find that our method can simulate human-chatbot dialogues with a high indistinguishability rate.

LLM Roleplay: Simulating Human-Chatbot Interaction

This paper presents PROTECT, a novel policy-driven organizational value taxonomy designed to enhance ethical compliance and trust within organizations. Drawing on established human value systems and leveraging large language models, PROTECT generates values tailored to organizational contexts and clusters them into a refined taxonomy. This taxonomy serves as the basis for creating a comprehensive dataset of compliance scenarios, each linked to specific values and paired with both compliant and non-compliant responses. By systematically varying value emphasis, we illustrate how different LLM personas emerge, reflecting diverse compliance behaviors. The dataset, directly grounded in the taxonomy, enables consistent evaluation and training of LLMs on value-sensitive tasks. While PROTECT offers a robust foundation for aligning AI systems with organizational standards, our experiments also reveal current limitations in model accuracy, highlighting the need for further improvements. Together, the taxonomy and dataset represent complementary, foundational contributions toward value-aligned AI in organizational settings.

PROTECT: Policy-Related Organizational Value Taxonomy for Ethical Compliance and Trust

As LLMs are increasingly used in global conversational settings, concerns remain about their ability to handle complex sociocultural contexts. This study evaluates LLMs’ empathetic understanding in Korean—a high-context language—using a pragmatics-based Discourse Completion Task (DCT) focused on interpretive judgment rather than generation. We constructed a dataset varying relational hierarchy, intimacy, and emotional valence, and compared responses from proprietary and open-source LLMs to those of Korean speakers. Most LLMs showed over-empathizing tendencies and struggled with ambiguous relational cues. Neither model size nor Korean fine-tuning significantly improved performance. While humans reflected relational nuance and contextual awareness, LLMs relied on surface strategies. These findings underscore LLMs’ limits in socio-pragmatic reasoning and introduce a scalable, culturally flexible framework for evaluating socially-aware AI.



Downloads

Next from ACL 2025

Mitigating Gender Bias in Job Ranking Systems Using Job Advertisement Neutrality

Similar lecture

Aligning Unstructured Paris Agreement Climate Plans with Sustainable Development Goals

.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 ACL 2025

Mitigating Gender Bias in Job Ranking Systems Using Job Advertisement Neutrality

Similar lecture

Aligning Unstructured Paris Agreement Climate Plans with Sustainable Development Goals

Downloads