Argument Mining (AM) involves the automatic identification of argument structure in natural language. Traditional AM methods rely on micro-structural features derived from the internal properties of individual Argumentative Discourse Units (ADUs). However, argument structure is shaped by a macro-structure capturing the functional interdependence among ADUs. This macro-structure consists of segments, where each segment contains ADUs that fulfill specific roles to maintain coherence within the segment (**local coherence**) and across segments (**global coherence**). This paper presents an approach that models macro-structure, capturing both local and global coherence to identify argument structures. Experiments on heterogeneous datasets demonstrate superior performance in both in-dataset and cross-dataset evaluations. The cross-dataset evaluation shows that macro-structure enhances transferability to unseen datasets.

Despite extensive research in Argument Mining (AM), the field faces significant challenges in limited reproducibility, difficulty in comparing systems due to varying task combinations, and a lack of interoperability caused by the heterogeneous nature of argumentation theory. These challenges are further exacerbated by the absence of dedicated tools, with most advancements remaining isolated research outputs rather than reusable systems. The oAMF (Open Argument Mining Framework) addresses these issues by providing an open-source, modular, and scalable platform that unifies diverse AM methods. Initially released with seventeen integrated modules, the oAMF serves as a starting point for researchers and developers to build, experiment with, and deploy AM pipelines while ensuring interoperability and allowing multiple theories of argumentation to co-exist within the same framework. Its flexible design supports integration via Python APIs, drag-and-drop tools, and web interfaces, streamlining AM development for research and industry setup, facilitating method comparison, and reproducibility.

The Open Argument Mining Framework (oAMF) addresses key challenges in argument mining research which still persist despite the field’s impressive growth. Researchers often face difficulties with cross-system comparisons, incompatible representation languages, and limited access to reusable tools. The oAMF introduces a standardised yet flexible architecture that enables seamless component benchmarking, rapid pipeline prototyping using elements from diverse research traditions, and unified evaluation methodologies that preserve theoretical compatibility. By reducing technical overhead, the framework allows researchers to focus on advancing core argument mining capabilities rather than reimplementing infrastructure, fostering greater collaboration at a time when computational reasoning is increasingly vital in the era of large language models.

Traditionally, argument mining research has approached the task of automatic identification of argument structures by using existing definitions of what constitutes an argument, while leaving the equally important matter of what does not qualify as an argument unaddressed. With the ability to distinguish between what is and what is not a natural language argument being at the core of argument mining as a field, it is interesting that no previous work has explored approaches to effectively select non-related propositions (i.e., propositions that are not connected through an argumentative relation, such as support or attack) that improve the data for learning argument mining tasks better. In this paper, we address the question of how to effectively sample non-related propositions from six different argument mining corpora belonging to different domains and encompassing both monologue and dialogue forms of argumentation. To that end, in addition to considering undersampling baselines from previous work, we propose three new sampling strategies relying on context (i.e., short/long) and the semantic similarity between propositions. Our results indicate that using more informed sampling strategies improves the performance, not only when evaluating models on their respective test splits, but also in the case of cross-domain evaluation.

While Large Language Models (LLMs) have demonstrated promising results on a range of reasoning benchmarks—particularly in formal logic, mathematical tasks, and Chain-of-Thought prompting—less is known about their capabilities in unconstrained natural language reasoning. Argumentative reasoning, a form of reasoning naturally expressed in language and central to everyday discourse, presents unique challenges for LLMs due to its reliance on context, implicit assumptions, and value judgments. This paper addresses a gap in the study of reasoning in LLMs by presenting the first large-scale evaluation of their unconstrained natural language reasoning capabilities based on natural language argumentation. The paper offers three contributions: (i) the formalisation of a new strategy designed to evaluate argumentative reasoning in LLMs: argument-component selection; (ii) the creation of the Argument Reasoning Tasks (ART) dataset, a new benchmark for argument-component selection based on argument structures for natural language reasoning; and (iii) an extensive experimental analysis involving four different models, demonstrating the limitations of LLMs on natural language reasoning tasks.

Measuring advances in argument mining is one of the main challenges in the area. Different theories of argument, heterogeneous annotations, and a varied set of argumentation domains make it difficult to contextualise and understand the results reported in different work from a general perspective. In this paper, we present ARIES, a general benchmark for Argument Relation Identification aimed at providing with a standard evaluation for argument mining research. ARIES covers the three different language modelling approaches: sequence and token modelling, and sequence-to-sequence-to-sequence alignment, together with the three main Transformer-based model architectures: encoder-only, decoder-only, and encoder-decoder. Furthermore, the benchmark consists of eight different argument mining datasets, covering the most common argumentation domains, and standardised with the same annotation structures. This paper provides a first comprehensive and comparative set of results in argument mining across a broad range of configurations to compare with, both advancing the state-of-the-art, and establishing a standard way to measure future advances in the area. Across varied task setups and architectures, our experiments reveal consistent challenges in cross-dataset evaluation, with notably poor results. Given the models’ struggle to acquire transferable skills, the task remains challenging, opening avenues for future research.

Argument mining (AM) involves the identification of argument relations (AR) between Argumentative Discourse Units (ADUs). The essence of ARs among ADUs is context-dependent and lies in maintaining a coherent flow of ideas, often centered around the relations between discussed entities, topics, themes or concepts. However, these relations are not always explicitly stated; rather, inferred from implicit chains of reasoning connecting the concepts addressed in the ADUs. While humans can infer such background knowledge, machines face challenges when the contextual cues are not explicitly provided. This paper leverages external resources, including WordNet, ConceptNet, and Wikipedia to identify semantic paths (knowledge paths) connecting the concepts discussed in the ADUs to obtain the implicit chains of reasoning. To effectively leverage these paths for AR prediction, we propose attention-based Multi-Network architectures. Various architecture are evaluated on the external resources, and the Wikipedia based configuration attains F-scores of 0.85, 0.84, 0.70, and 0.87, respectively, on four diverse datasets, showing strong performance over the baselines.

This work presents an approach decomposing propositions into four functional components and identify the patterns linking those components to determine argument structure. The entities addressed by a proposition are target concepts and the features selected to make a point about the target concepts are aspects. A line of reasoning is followed by providing evidence for the points made about the target concepts via aspects. Opinions on target concepts and opinions on aspects are used to support or attack the ideas expressed by target concepts and aspects. The relations between aspects, target concepts, opinions on target concepts and aspects are used to infer the argument relations. Propositions are connected iteratively to form a graph structure. The approach is generic in that it is not tuned for a specific corpus and evaluated on three different corpora from the literature: AAEC, AMT, US2016G1tv and achieved an F score of 0.79, 0.77 and 0.64, respectively.