Austria

Through manual annotation or automated argument mining processes, arguments can be represented not only as text, but also in structured formats like graphs. When searching for relevant arguments, this additional information about the relationship between their elementary units allows for the formulation of fine-grained structural constraints by using graphs as queries. Then, a retrieval can be performed by computing the similarity between the query and all available arguments. Previous works employed Graph Edit Distance (GED) algorithms such as A* search to compute mappings between nodes and edges for determining the similarity, which is rather expensive. In this paper, we propose an alternative based on Vision Transformers where arguments are rendered as images to obtain dense embeddings. We propose multiple space-filling visualizations and evaluate the retrieval performance of the vision-based approach against an existing A* search-based method. We find that our technique runs orders of magnitude faster than A* search and scales well on larger argument graphs while achieving competitive results.

ACL 2025

Retrieving Argument Graphs Using Vision Transformers

**Welcome to ACL-IJCNLP 2021!**

The great event is jointly organized by the Association for Computational Linguistics (ACL) and Asian Federation of Natural Language Processing (AFNLP). 

As in previous years, the program of the conference includes a poster session, tutorials, workshops and demonstrations in addition to the main conference.


We were able to keep the registration fees similar to those charged for the virtual ACL 2020. The one fee allows attendance at the main conference and any/all tutorials and workshops. These fees would be $125 Regular Early and $175 Regular Late; $50 Student Early and $75 Student Late. Early registration closes at midnight July 11, 2021 (Eastern Daylight Time).

**Reminder:** It is ACL’s policy that at least one author of each accepted paper (including ACL Finding papers) must register for the conference.

**Reminder2:** Underline site will open closer to the event. If you already registered you will receive access detail

Registration is now open

ACL-IJCNLP 2021

The great event is jointly organized by the Association for Computational Linguistics (ACL) and Asian Federation of Natural Language Processing (AFNLP).

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)

Online discussions can either bridge differences through constructive dialogue or amplify divisions through destructive interactions. This paper proposes a computational approach to analyze dialogical relation patterns in YouTube comments, offering a fine-grained framework for controversy detection, enabling also analysis of individual contributions. Our experiments demonstrate that shallow learning methods, when equipped with these theoretically-grounded features, consistently outperform more complex language models in characterizing discourse quality at both comment-pair and conversation-chain levels. Ablation studies confirm that divisive rhetorical techniques serve as strong predictors of destructive communication patterns. This work advances understanding of how communicative choices shape online discourse, moving beyond engagement metrics toward nuanced examination of constructive versus destructive dialogue patterns.

Old but Gold: LLM-Based Features and Shallow Learning Methods for Fine-Grained Controversy Analysis in YouTube Comments

Implicit premise is central to argumentative coherence and faithfulness, yet remain elusive in traditional single-pass computational models. We introduce a multi-agent framework that casts implicit premise recovery as a dialogic reasoning task between two LLM agents. Through structured rounds of debate, agents critically evaluate competing premises and converge on the most contextually appropriate interpretation. Evaluated on a controlled binary classification benchmark for premise selection, our approach achieves state-of-the-art accuracy, outperforming both neural baselines and single-agent LLMs. We find that accuracy gains stem not from repeated generation, but from agents refining their predictions in response to opposing views. Moreover, we show that forcing models to defend assigned stances degrades performance—engendering rhetorical rigidity to flawed reasoning. These results underscore the value of interactive debate in revealing pragmatic components of argument structure.

Multi-Agent LLM Debate Unveils the Premise Left Unsaid

Argument Mining (AM) aims at detecting argumentation structures (i.e., premises and claims linked by attack and support relations) in text. A natural application domain is political debates, where uncovering the hidden dynamics of a politician's argumentation strategies can help the public to identify fallacious and propagandist arguments. Despite the few approaches proposed in the literature to apply AM to political debates, this application scenario is still challenging, and, more precisely, concerning the task of predicting the relation holding between two argument components. Most of AM relation prediction approaches only consider the textual content of the argument component to identify and classify the argumentative relation holding among them (i.e., support, attack), and they mostly ignore the structural knowledge that arises from the overall argumentation graph. In this paper, we propose to address the relation prediction task in AM by combining the structural knowledge provided by a Knowledge Graph Embedding Model with the contextual knowledge provided by a fine-tuned Language Model. Our experimental setting is grounded on a standard AM benchmark of televised political debates of the US presidential campaigns from 1960 to 2020. Our extensive experimental setting demonstrates that integrating these two distinct forms of knowledge (i.e., the textual content of the argument component and the structural knowledge of the argumentation graph) leads to novel pathways that outperform existing approaches in the literature on this benchmark and enhance the accuracy of the predictions.

Leveraging Graph Structural Knowledge to Improve Argument Relation Prediction in Political Debates

Given the recent success of LLMs across different NLP tasks, their usability for data annotation has become a promising area of research. In this work, we investigate to what extent LLMs can be used as annotators for argument components and their semantic types in German tweets through a series of experiments combining different models and prompt configurations. Each prompt is constructed from modular components, such as class definitions or contextual information. Our results suggest that LLMs can indeed perform argument annotation, particularly of semantic argument types, if provided with precise class definitions. However, a fine-tuned BERT baseline remains a strong contender, often matching or exceeding LLM performance. These findings highlight the importance of considering not only model performance, but also ecological and financial costs when defining an annotation workflow.

On Integrating LLMs Into an Argument Annotation Workflow

Legal reasoning remains one of the most complex and nuanced domains for AI, with current tools often lacking transparency and domain adaptability. While recent advances in large language models (LLMs) offer new opportunities for legal analysis, their ability to structure and interpret judicial argumentation remains unexplored. We address this gap by proposing a structured framework for AI-assisted legal reasoning, centered on argumentative analysis. In this work, we use GPT-4o for discourse-level and semantic analysis to identify argumentative units and classify them according to Philippe Bobbitt’s six constitutional modalities of legal reasoning. We apply this framework to legal rulings from the Italian Court of Cassation. Our experimental findings indicate that LLM-based tools can effectively augment and streamline legal practice, by e.g. preprocessing the legal texts under scrutiny; still, the limited performance of the state-of-the-art generative model tested indicates significant room for progress in human-AI collaboration in the legal domain.

Argumentative Analysis of Legal Rulings: A Structured Framework Using Bobbitt’s Typology

Reviews are valuable resources for customers making purchase decisions in online shopping. However, it is impractical for customers to go over the vast number of reviews and manually conclude the prominent opinions, which prompts the need for automated opinion summarization systems. Previous approaches, either extractive or abstractive, face challenges in automatically producing grounded aspect-centric summaries. In this paper, we propose a novel summarization system that not only captures predominant opinions from an aspect perspective with supporting evidence, but also adapts to varying domains without relying on a pre-defined set of aspects. Our proposed framework, ASESUM, summarizes viewpoints relevant to the critical aspects of a product by extracting aspect-centric arguments and measuring their \textit{salience} and \textit{validity}. We conduct experiments on a real-world dataset to demonstrate the superiority of our approach in capturing diverse perspectives of the original reviews compared to new and existing methods.

Aspect-Based Opinion Summarization with Argumentation Schemes

In assessing argument strength, the notions of what makes a good argument are manifold. With the broader trend towards treating subjectivity as an asset and not a problem in NLP, new dimensions of argument quality are studied. Although studies on individual subjective features like personal stories exist, there is a lack of large-scale analyses of the relation between these features and argument strength. To address this gap, we conduct regression analysis to quantify the impact of subjective factors – emotions, storytelling, and hedging on two standard datasets annotated for objective argument quality and subjective persuasion. As such, our contribution is twofold: at the level of contributed resources, as there are no datasets annotated with all studied dimensions, this work compares and evaluates automated annotation methods for each subjective feature. At the level of novel insights, our regression analysis uncovers different patterns of impact of subjective features on the two facets of argument strength encoded in the datasets. Our results show that storytelling and hedging have contrasting effects on objective and subjective argument quality, while the influence of emotions depends on their rhetoric utilization rather than the domain.

Investigating Subjective Factors of Argument Strength: Storytelling, Emotions, and Hedging

Television debates play a key role in shaping public opinion, however, the rapid exchange of viewpoints in these settings often makes it difficult to perceive the underlying nature of the discussion. While there exist several debate visualisation techniques, to the best of our knowledge, none of them emphasise the argumentative dynamics in particular. With DebArgVis, we present a new interactive debate visualisation tool that leverages data annotated with argumentation structures to demonstrate how speaker interactions unfold over time, enabling users to deepen their comprehension of the debate.

DebArgVis: An Interactive Visualisation Tool for Exploring Argumentative Dynamics in Debate

Being aware of frames, i.e., the aspect-based grouping of arguments, is crucial in applications that build upon a corpus of arguments, allowing, among others, biases and filter bubbles to be mitigated. However, manually identifying and naming these frames can be time-consuming and therefore not feasible for larger datasets. Within this work, we present a sequential three-step pipeline for automating this task in a data-driven manner. After embedding the arguments, we apply clustering algorithms for identifying the frames and subsequently, utilize methods from the field of cluster labeling to name the frames. The proposed approach is tailored towards the requirements of practical applications where arguments may not be easily split into their argumentative units and hence can belong to more than one frame. Performing a component-wise evaluation, we determine the best-performing configuration of the pipeline. Our results indicate that frames should be identified by performing overlapping and not exclusive clustering and the naming of frames can be accomplished best by extracting aspect terms and weighting them with c-TF-IDF.

Automatic Identification and Naming of Overlapping and Topic-specific Argumentation Frames

A crucial task when analyzing arguments is to determine their quality. Especially when you have to choose from a large number of suitable arguments, the determination of a reliable argument quality value is of great benefit. Probably the best-known model for determining such an argument quality value was developed in IBM's Project Debater and made available to the research community free of charge via an API. In fact, the model was never open and the API is no longer available. In this paper, IBM's model is reproduced using the freely available training data and the description in the corresponding publication. Our reproduction achieves similar results on the test data as described in the original publication. Further, the predicted quality scores of reproduction and original show a very high correlation (Pearson's $r=0.9$) on external data.

Downloads

Next from ACL 2025

Old but Gold: LLM-Based Features and Shallow Learning Methods for Fine-Grained Controversy Analysis in YouTube Comments

Similar lecture

Employing Argumentation Knowledge Graphs for Neural Argument Generation