As the use of AI in society grows, addressing emerging biases is essential to prevent systematic discrimination. Several bias detection methods have been proposed, but, with few exceptions, these tend to ignore transparency. Instead, interpretability and explainability are core requirements for algorithmic fairness, even more so than for other algorithmic solutions, given the human-oriented nature of fairness. We present ABIDE (Argumentative BIas detection by DEbate), a novel framework that structures bias detection transparently as debate, guided by an underlying argument graph as understood in (formal and computational) argumentation. The arguments are about the success chances of groups in local neighbourhoods and the significance of these neighbourhoods. We evaluate ABIDE experimentally and demonstrate its strengths in performance against an argumentative baseline.

Assumption‐Based Argumentation (ABA) is a powerful structured argumentation formalism, but exact computation of extensions under stable semantics is intractable for large frameworks. We present the first Graph Neural Network (GNN) approach to approximate credulous acceptance in ABA. To leverage GNNs, we model ABA frameworks via a dependency graph representation encoding assumptions, claims and rules as nodes, with heterogeneous edge labels distinguishing support, derive and attack relations. We propose two GNN architectures—ABAGCN and ABAGAT—that stack residual heterogeneous convolution or attention layers, respectively, to learn node embeddings. Our models are trained on the ICCMA 2023 benchmark, augmented with synthetic ABAFs, with hyperparameters optimised via Bayesian search. Empirically, both ABAGCN and ABAGAT outperform a state‐of‐the‐art GNN baseline that we adapt from the abstract argumentation iterature, achieving a node‐level F1 score of up to 0.71 on the ICCMA instances. Finally, we develop a sound polynomial time extension‐reconstruction algorithm driven by our predictor: it reconstructs stable extensions with F1 above 0.85 on small ABAFs and maintains an F1 of about 0.58 on large frameworks. Our work opens new avenues for scalable approximate reasoning in structured argumentation.

ABA Learning is a recent approach for obtaining Assumption-based Argumentation (ABA) frameworks by reasoning with transformation rules from background knowledge and positive/negative examples of concepts of interest. ABA Learning relies on credulous reasoning under a specific semantic notion of extensions for ABA, namely that of stable extensions. In this paper, we newly frame the problem in terms of credulous reasoning under any semantic notion of extensions for ABA. Focusing on admissible, complete, grounded, preferred as well as stable extensions, we present X-ABALearn, a novel parametric algorithm (with X any ABA semantics) based on variants of the transformation rules of ABA Learning and an implementation thereof in Answer Set Programming. Finally, we explore the use of (our implementation of) X-ABALearn on several learning problems, including tabular data and beyond.

Large language models (LLMs) exhibit strong general capabilities, but their deployment in high-stakes domains is hindered by their opacity and unpredictability. Recent work has taken meaningful steps towards addressing these issues by augmenting LLMs with post-hoc reasoning based on computational argumentation, providing faithful explanations and enabling users to contest incorrect decisions. However, this paradigm is limited to pre-defined binary choices and only supports local contestation for specific instances, leaving the underlying decision logic unchanged and prone to repeated mistakes. In this paper, we introduce ArgEval, a framework that shifts from instance-specific reasoning to structured evaluation of general decision options. Rather than mining arguments solely for individual cases, ArgEval systematically maps task-specific decision spaces, builds corresponding option ontologies, and constructs general argumentation frameworks (AFs) for each option. These frameworks can then be instantiated to provide explainable recommendations for specific cases while still supporting global contestability through modification of the shared AFs. We investigate the effectiveness of ArgEval on treatment recommendation for glioblastoma, an aggressive brain tumour, and show that it can produce explainable guidance aligned with clinical practice.

Contestable AI requires that AI-driven decisions align with given preferences. Various types of argumentation frameworks have been shown to support forms of contestability. In this paper we focus on the little-studied Edge-Weighted Quantitative Bipolar Argumentation Frameworks (EW-QBAFs), where arguments have a base score as in QBAFs but attacks and supports (edges) are weighted. After generalising gradual semantics and properties thereof from QBAFs to EW-QBAFs, we introduce the contestability problem for EW-QBAFs, which asks how to modify edge weights to achieve a desired strength for a specific topic argument. To address this problem, we propose gradient-based relation attribution explanations (G-RAEs), which quantify the sensitivity of the topic argument's strength to changes in individual edge weights, thus providing interpretable guidance for weight adjustments towards contestability. Building on G-RAEs, we develop a heuristic algorithm that progressively adjusts the edge weights to attain the desired strength. We evaluate our approach experimentally on synthetic EW-QBAFs that simulate the structural characteristics of personalised recommender systems and multi-layer perceptrons, demonstrating that it can support contestability effectively.

We propose ProtoArgNet, a novel interpretable deep neural architecture for image classification in the spirit of prototypical-part-learning as found, e.g., in ProtoPNet. While earlier approaches associate every class with multiple prototypical-parts, ProtoArgNet uses super-prototypes that combine prototypical-parts into a unified class representation. This is done by combining local activations of prototypes in an MLP-like manner, enabling the localization of prototypes and learning (non-linear) spatial relationships among them. By leveraging a form of argumentation, ProtoArgNet is capable of providing both supporting (i.e. `this looks like that') and attacking (i.e. `this differs from that') explanations. We demonstrate on several datasets that ProtoArgNet outperforms state-of-the-art prototypical-part-learning approaches. Moreover, the argumentation component in ProtoArgNet is customisable to the user's cognitive requirements by a process of sparsification, which leads to more compact explanations compared to state-of-the-art approaches.

Large Language Models (LLMs) possess vast amounts of knowledge within their parameters, prompting research into methods for locating and editing this knowledge. Previous work has largely focused on locating entity-related (often single-token) facts in smaller models. However, several key questions remain unanswered: (1) How can we effectively locate query-relevant neurons in contemporary autoregressive LLMs, such as Llama and Mistral? (2) How can we address the challenge of long-form text generation? (3) Are there localized knowledge regions in LLMs? In this study, we introduce Query-Relevant Neuron Cluster Attribution (QRNCA), a novel architecture-agnostic framework capable of identifying query-relevant neurons in LLMs. QRNCA allows for the examination of long-form answers beyond triplet facts by employing the proxy task of multi-choice question answering. To evaluate the effectiveness of our detected neurons, we build two multi-choice QA datasets spanning diverse domains and languages. Empirical evaluations demonstrate that our method outperforms baseline methods significantly. Further, analysis of neuron distributions reveals the presence of visible localized regions, particularly within different domains. Finally, we show potential applications of our detected neurons in knowledge editing and neuron-based prediction.

The profusion of knowledge encoded in large language models (LLMs) and their ability to apply this knowledge zero-shot in a range of settings makes them promising candidates for use in decision-making. However, they are currently limited by their inability to provide outputs which can be faithfully explained and effectively contested to correct mistakes. In this paper, we attempt to reconcile these strengths and weaknesses by introducing argumentative LLMs (ArgLLMs), a method for augmenting LLMs with argumentative reasoning. Concretely, ArgLLMs construct argumentation frameworks, which then serve as the basis for formal reasoning in support of decision-making. The interpretable nature of these argumentation frameworks and formal reasoning means that any decision made by ArgLLMs may be explained and contested. We evaluate ArgLLMs’ performance experimentally in comparison with state-of-the-art techniques, in the context of the decision-making task of claim verification. We also define novel properties to characterise contestability and assess ArgLLMs formally in terms of these properties.

Relation-based Argument Mining (RbAM) is the process of automatically determining agreement (support) and disagreement (attack) relations amongst textual arguments (in the binary prediction setting), or neither relation (in the ternary prediction setting). As the number of platforms supporting online debate increases, the need for RbAM becomes ever more urgent, especially in support of downstream tasks. RbAM is a challenging classification task, with existing state-of-the-art methods, based on Language Models (LMs), failing to perform satisfactorily across different datasets. In this paper, we show that general-purpose Large LMs (LLMs), appropriately primed and prompted, can significantly outperform the best performing (RoBERTa-based) baseline. Specifically, we experiment with two open-source LLMs (Llama-2 and Mistral) and with GPT-3.5-turbo on several datasets for (binary and ternary) RbAM, as well as with GPT-4o-mini on samples (to limit costs) from the datasets.

AI-driven tools for healthcare are widely acknowledged as potentially beneficial to health practitioners and patients, e.g. the QCancer regression tool for cancer risk prediction. However, for these tools to be trusted, they need to be supplemented with explanations. We examine how explanations’ content and format affect user comprehension and trust when explaining QCancer’s predictions. Regarding content, we deploy the SHAP and Occlusion-1 explanation methods. Regarding format, we present SHAP explanations, conventionally, as charts (SC) and Occlusion-1 explanations as charts (OC) as well as text (OT), to which their simpler nature lends itself. We conduct experiments with two sets of stakeholders: the general public (representing patients) and medical students (representing healthcare practitioners). Our experiments showed higher subjective comprehension and trust for Occlusion-1 over SHAP explanations based on content. However, when controlling for format, only OT outperformed SC, suggesting this trend is driven by preferences for text. Other findings corroborated that explanation format, rather than content, is often the critical factor.

Today’s online platforms, e.g. in e-commerce, often offer users numerous competing options in single product categories, e.g. televisions or watches, making it difficult for the users to identify the best option to suit their preferences. To ease this process, many platforms provide users with simple scores resulting from the aggregation of other users’ reviews. However, these scoring systems may oversimplify the underlying information and lack explanatory context. Our main contribution in this demonstration paper is a novel online system for aggregating customer reviews and explaining the aggregation to users.The system operates through a multi-stage pipeline: it first applies novel automatic ontology extraction methods using BERT or Large Language Models to identify key aspects from customer reviews, then constructs support and attack relations between these aspects using Argumentative Dialogical Agents (ADAs), an existing methodology for generating argumentative analyses of aspects. Finally, it generates ontology-driven, explainable aggregations of the reviews. We evaluate the performance of our system (which we call ADA-X) on the Amazon and Disneyland review datasets, focusing the ontology quality using the LLM-as-a-judge method and aggregation performance against the original Amazon and Disneyland ratings. The demonstration is available at https://ada-x.co.uk/.

Judgmental forecasting employs human opinions to make predictions about future events, rather than exclusively historical data as in quantitative forecasting. When these opinions form an argumentative structure around forecasts, it is useful to study the properties of the forecasts from an argumentative perspective. In this paper, we advocate and formally define a property of argumentative coherence, which, in essence, requires that a forecaster’s reasoning is coherent with their forecast. We then conduct three evaluations with our notion of coherence. First, we assess the impact of enforcing coherence on human forecasters as well as on Large Language Model (LLM)-based forecasters, given that they have recently shown to be competitive with human forecasters. In both cases, we show that filtering out incoherent predictions improves forecasting accuracy consistently, supporting the practical value of coherence in both human and LLM-based forecasting. Then, via crowd-sourced user experiments, we show that, despite its apparent intuitiveness and usefulness, users do not generally align with this coherence property. This points to the need to integrate, within argumentation-based judgmental forecasting, mechanisms to filter out incoherent opinions before obtaining group forecasting predictions.

Workforce management is a complex problem involving the optimisation of the makespan and travel distance required for a team of operators to complete a set of jobs, using a set of instruments. A crucial challenge in workforce management is accommodating changes at execution time so that explanations are provided to all stakeholders involved. Here, we show that, by understanding workforce management as abstract argumentation in an industrial application, we can accommodate change and obtain faithful explanations. We show, with a user study, that our tool and explanations lead to faster and more accurate problem solving than conventional manual approaches.

Composites are amongst the most important materials manufactured today, as evidenced by their use in countless applications. In order to establish the suitability of composites in specific applications, finite element (FE) modelling, a numerical method based on partial differential equations, is the industry standard for assessing their mechanical properties. However, FE modelling is exceptionally costly from a computational viewpoint, a limitation which has led to efforts towards applying AI models to this task. However, in these approaches: the chosen model architectures were rudimentary, feed-forward neural networks giving limited accuracy; the studies focused on predicting elastic mechanical properties, without considering material strength limits; and the models lacked transparency, hindering trustworthiness by users. In this paper, we show that convolutional neural networks (CNNs) equipped with methods from explainable AI (XAI) can be successfully deployed to solve this problem. Our approach uses customised CNNs trained on a dataset we generate using transverse tension tests in FE modelling to predict composites’ mechanical properties, i.e., Young’s modulus and yield strength. We show empirically that our approach achieves high accuracy, outperforming a baseline, ResNet-34, in estimating the mechanical properties. We then use SHAP and Integrated Gradients, two post-hoc XAI methods, to explain the predictions, showing that the CNNs use the critical geometrical features that influence the composites’ behaviour, thus allowing engineers to verify that the models are trustworthy by representing the science of composites.

Modular object-centric representations are essential for human-like reasoning but are challenging to obtain under spatial ambiguities, e.g. due to occlusions and view ambiguities. However, addressing challenges presents both theoretical and practical difficulties. We introduce a novel multi-view probabilistic approach that aggregates view-specific slots to capture invariant content information while simultaneously learning disentangled global viewpoint-level information. Unlike prior single-view methods, our approach resolves spatial ambiguities, provides theoretical guarantees for identifiability, and requires no viewpoint annotations. Extensive experiments on standard benchmarks and novel complex datasets validate our method’s robustness and scalability.

Counterfactual explanations (CXs) are widely recognised as an essential technique for providing recourse recommendations for AI models. However, it is not obvious how to determine CXs in model multiplicity scenarios, where equally performing but different models can be obtained for the same task. In this paper, we propose novel qualitative and quantitative definitions of CXs based on explicit, nested quantification over (groups) of model decisions. We also study properties of these notions and identify decision problems of interest therefor. While our CXs are broadly applicable, in this paper we instantiate them within computational argumentation where model multiplicity naturally emerges, e.g. with incomplete and case-based argumentation frameworks. We then illustrate the suitability of our CXs for model multiplicity in legal and healthcare contexts, before analysing the complexity of the associated decision problems.

We introduce a notion of conditional independence in (flat) assumption-based argumentation (ABA), where independence between (sets of) assumptions amounts to the presence of information about one set of assumptions not impacting the acceptability of another. We study general properties, computational complexity, and the relation to independence in abstract argumentation. In light of the high computational complexity of deciding independence, we introduce sound methods for checking independence in polynomial time via two different routes: the first utilizes the strongly connected components (SCCs) of the instantiated abstract argumentation framework; the second exploits the structure of the ABA framework directly. Along the way, we introduce the notion of SCC-recursiveness for ABA.

If legal professionals are to use Large Language Models (LLMs), LLMs must provide explanations to empower safe decision making. Argumentative LLMs (ArgLLMs) were recently introduced to produce and explain decisions in the form of claim verifications; they provide explanations that faithfully match their underlying reasoning and are contestable by human users. In this paper, ArgLLMs are applied to two COLIEE 2025 legal entailment tasks: task 4, which involves determining whether relevant statute articles entail a legal hypothesis, and the pilot task, which involves determining whether tort case information entails the conclusion that the tort case was affirmed by the judge. We perform an ablation study to assess how several modifications to ArgLLMs affect accuracy in the two tasks. We also compare the performance of our variants of ArgLLMs against two baselines, one involving a single zero-shot Chain of Thought (CoT) prompt and another involving a pairwise comparison of supporting and attacking arguments. Our experiments show that ArgLLMs are more accurate than over half of all official submissions to task 4 and the pilot task of the COLIEE competition. Moreover, Arg-LLMs produce accuracy scores similar to the CoT baseline, while also providing the benefit of faithful and contestable explanations behind the decision made. For task 4, we also conducted a pilot study where legal experts reviewed 20 generated explanations and found the arguments on the correct side of the debate to be both sound and faithful to the legal context provided in 17 of them. Repository link: https://github.com/charlieblindsay/arg-llms

Gradual semantics (GS) have demonstrated great potential in argumentation, in particular for deploying quantitative bipolar argumentation frameworks (QBAFs) in a number of real-world settings, from judgmental forecasting to explainable AI. In this paper, we provide a novel methodology for obtaining GS for statement graphs, a form of structured argumentation framework, where arguments and relations between them are built from logical statements. Our methodology differs from existing approaches in the literature in two main ways. First, it naturally accommodates incomplete information, so that arguments with partially specified premises can play a meaningful role in the evaluation. Second, it is modularly defined to leverage on any GS for QBAFs. We also define a set of novel properties for our GS and study their suitability alongside a set of existing properties (adapted to our setting) for two instantiations of our GS, demonstrating their advantages over existing approaches.

In computational argumentation, gradual semantics are fine-grained alternatives to extension-based and labelling-based semantics. They ascribe a dialectical strength to (components of) arguments sanctioning their degree of acceptability. Several gradual semantics have been studied for abstract, bipolar and quantitative bipolar argumentation frameworks (QBAFs), as well as, to a lesser extent, for some forms of structured argumentation. However, this has not been the case for assumption-based argumentation (ABA), despite it being a popular form of structured argumentation with several applications where gradual semantics could be useful. In this paper, we fill this gap and propose a family of novel gradual semantics for equipping assumptions, which are the core components in ABA frameworks, with dialectical strengths. To do so, we use bipolar set-based argumentation frameworks as an abstraction of (potentially non-flat) ABA frameworks and generalise state-of-the-art modular gradual semantics for QBAFs. We show that our gradual ABA semantics satisfy suitable adaptations of desirable properties of gradual QBAF semantics, such as balance and monotonicity. We also explore an argument-based approach that leverages established QBAF modular semantics directly, and use it as baseline. Finally, we conduct experiments with synthetic ABA frameworks to compare our gradual ABA semantics with its argument-based counterpart and assess convergence.

We introduce Gradual Abstract Argumentation for Case-Based Reasoning (Gradual AA-CBR), a data-driven, neurosymbolic classification model in which the outcome is determined by an argumentation debate structure that is learned simultaneously with neural-based feature extractors. Each argument in the debate is an observed case from the training data, favouring their labelling. Cases attack or support those with opposing or agreeing labellings, with the strength of each argument and relationship learned through gradient-based methods. This argumentation debate structure provides human-aligned reasoning, improving model interpretability compared to traditional neural networks (NNs). Unlike the existing purely symbolic variant, Abstract Argumentation for Case-Based Reasoning (AA-CBR), Gradual AA-CBR is capable of multi-class classification, automatic learning of feature and data point importance, assigning uncertainty values to outcomes, using all available data points, and does not require binary features. We show that Gradual AA-CBR performs comparably to NNs whilst significantly outperforming existing AA-CBR formulations.

Over the last decade, as we rely more on deep learning technologies to make critical decisions, concerns regarding their safety, reliability and interpretability have emerged. We introduce a novel Neural Argumentative Learning (NAL) architecture that integrates Assumption-Based Argumentation (ABA) with deep learning for image analysis. Our architecture consists of neural and symbolic components. The former segments and encodes images into facts using object-centric learning, while the latter applies ABA learning to develop ABA frameworks enabling predictions with images. Experiments on synthetic data show that the NAL architecture can be competitive with a state-of-the-art alternative.

Test-time scaling improves large language models' (LLMs) performance by allocating more compute budget during inference. To achieve this, existing methods often require intricate modifications to prompting and sampling strategies. In this work, we introduce representation consistency (RC), a test-time scaling method for aggregating answers drawn from multiple candidate responses of an LLM regardless of how they were generated, including variations in prompt phrasing and sampling strategy. RC enhances answer aggregation by not only considering the number of occurrences of each answer in the candidate response set, but also the consistency of the model's internal activations while generating the set of responses leading to each answer. These activations can be either dense (raw model activations) or sparse (encoded via pretrained sparse autoencoders). Our rationale is that if the model's representations of multiple responses converging on the same answer are highly variable, this answer is more likely to be the result of incoherent reasoning and should be down-weighted during aggregation. Importantly, our method only uses cached activations and lightweight similarity computations and requires no additional model queries. Through experiments with four open-source LLMs and four reasoning datasets, we validate the effectiveness of RC for improving task performance during inference, with consistent accuracy improvements (up to 4\%) over strong test-time scaling baselines. We also show that consistency in the sparse activation signals aligns well with the common notion of coherent reasoning.

Abstract We introduce a novel method to aggregate bipolar argumentation frameworks expressing opinions of different parties in debates. We use Bipolar Assumption-based Argumentation (ABA) as an all-encompassing formalism for bipolar argumentation under different semantics. By leveraging on recent results on judgement aggregation in social choice theory, we prove several preservation results for relevant properties of bipolar ABA using quota and oligarchic rules. Specifically, we prove (positive and negative) results about the preservation of conflict-free, closed, admissible, preferred, complete, set-stable, well-founded and ideal extensions in bipolar ABA, as well as the preservation of acceptability, acyclicity and coherence for individual assumptions. Finally, we illustrate our methodology and results in the context of a case study on opinion aggregation for the treatment of long COVID patients.

The aggregation of online reviews is one of the dominant methods of quality control for users in various domains, from retail to entertainment. Consequently, explainable aggregation of reviews is increasingly sought-after. We introduce quantitative argumentation technology to this setting, towards automatically generating reasoned review aggregations equipped with dialogical explanations. To this end, we define a novel form of argumentative dialogical agent (ADA), using ontologies to harbour information from reviews into argumentation frameworks. These agents may then be evaluated with a quantitative argumentation semantics and used to mediate the generation of dialogical explanations for item recommendations based on the reviews. We show how to deploy ADAs in three different contexts in which argumentation frameworks are mined from text, guided by ontologies. First, for hotel recommendations, we use a human-authored ontology and exemplify the potential range of dialogical explanations afforded by ADAs. Second, for movie recommendations, we empirically evaluate an ADA based on a bespoke ontology (extracted semi-automatically, by natural language processing), by demonstrating that its quantitative evaluations, which are shown to satisfy desirable theoretical properties, are comparable with those on a well-known movie review aggregation website. Finally, for product recommendation in e-commerce, we use another bespoke ontology (extracted fully automatically, by natural language processing , from a website's reviews) to construct an ADA which is then empirically evaluated favourably against review aggregations from the website.

Assumption-based Argumentation (ABA) is a well-known structured argumentation formalism, whereby arguments and attacks between them are drawn from rules, defeasible assumptions and their contraries. A common restriction imposed on ABA frameworks (ABAFs) is that they are flat, i.e. each of the defeasible assumptions can only be assumed, but not derived. While it is known that flat ABAFs can be translated into abstract argumentation frameworks (AFs) as proposed by Dung, no translation exists from general, possibly non-flat ABAFs into any kind of abstract argumentation formalism. In this paper, we close this gap and show that bipolar AFs (BAFs) can instantiate general ABAFs. To this end we develop suitable, novel BAF semantics which borrow from the notion of deductive support. We investigate basic properties of our BAFs, including computational complexity, and prove the desired relation to ABAFs under several semantics.

Neural networks (NNs) can learn to rely on spurious signals in the training data, leading to poor generalisation. Recent methods tackle this problem by training NNs with additional ground-truth annotations of such signals. These methods may, however, let spurious signals re-emerge in deep convolutional NNs (CNNs). We propose Targeted Activation Penalty (TAP), a new method tackling the same problem by penalising activations to control the re-emergence of spurious signals in deep CNNs, while also lowering training times and memory usage. In addition, ground-truth annotations can be expensive to obtain. We show that TAP still works well with annotations generated by pre-trained models as effective substitutes of ground-truth annotations. We demonstrate the power of TAP against two state-of-the-art baselines on the MNIST benchmark and on two clinical image datasets, using four different CNN architectures.

As deep neural models in NLP become more complex, and as a consequence opaque, the necessity to interpret them becomes greater. A burgeoning interest has emerged in rationalizing explanations to provide short and coherent justifications for predictions. In this position paper, we advocate for a formal framework for key concepts and properties about rationalizing explanations to support their evaluation systematically. We also outline one such formal framework, tailored to rationalizing explanations of increasingly complex structures, from free-form explanations to deductive explanations, to argumentative explanations (with the richest structure). Focusing on the automated fact verification task, we provide illustrations of the use and usefulness of our formalization for evaluating explanations, tailored to their varying structures.

Most existing computational tools for assumption-based argumentation (ABA) focus on so-called flat frameworks, disregarding the more general case. Here, we study an instantiation-based approach for reasoning in possibly non-flat ABA. For complete-based semantics, an approach of this kind was recently introduced, based on a semantics-preserving translation between ABA and bipolar argumentation frameworks (BAFs). Admissible semantics, however, require us to consider an extension of BAFs which also makes use of premises of arguments (pBAFs). We explore basic properties of pBAFs which we require as a theoretical underpinning for our proposed instantiation-based solver for non-flat ABA under admissible semantics. As our empirical evaluation shows, depending on the ABA instances, the instantiation-based solver is competitive against an ASP-based approach implemented in the style of state-of-the-art solvers for hard argumentation problems.

The field of explainable AI has grown exponentially in recent years. Within this landscape, argumentation frameworks have shown to be helpful abstractions of some AI models towards providing explanations thereof. While existing work on argumentative explanations and their properties has focused on static settings, we focus on dynamic settings whereby the (AI models underpinning the) argumentation frameworks need to change. Specifically, for a number of notions of explanations drawn from abstract argumentation frameworks under extension-based semantics, we address the following questions: (1) Are explanations robust to extension-preserving changes, in the sense that they are still valid when the changes do not modify the extensions? (2) If not, are these explanations pseudo-robust in that can be tractably updated? In this paper, we frame these questions formally. We consider robustness and pseudo-robustness w.r.t. ordinary and strong equivalence and provide several results for various extension-based semantics.

Assumption-based Argumentation (ABA) is advocated as a unifying formalism for various forms of non-monotonic reasoning, including logic programming. It allows capturing defeasible knowledge, subject to argumentative debate. While, in much existing work, ABA frameworks are given up-front, in this paper we focus on the problem of automating their learning from background knowledge and positive/negative examples. Unlike prior work, we newly frame the problem in terms of brave reasoning under stable extensions for ABA. We present a novel algorithm based on transformation rules (such as Rote Learning, Folding, Assumption Introduction and Fact Subsumption) and an implementation thereof that makes use of Answer Set Programming. Finally, we compare our technique to state-of-the-art ILP systems that learn defeasible knowledge.

The extraction of object-centric representations for downstream tasks is an emerging area of research. Learning grounded representations of objects that are guaranteed to be stable and invariant promises robust performance across different tasks and environments. Slot Attention (SA) learns object-centric representations by assigning objects to *slots*, but presupposes a *single* distribution from which all slots are randomly initialised. This results in an inability to learn *specialized* slots which bind to specific object types and remain invariant to identity-preserving changes in object appearance. To address this, we present *Conditional Slot Attention* (CoSA) using a novel concept of *Grounded Slot Dictionary* (GSD) inspired by vector quantization. Our proposed GSD comprises (i) canonical object-level property vectors and (ii) parametric Gaussian distributions, which define a prior over the slots. We demonstrate the benefits of our method in multiple downstream tasks such as scene generation, composition, and task adaptation, whilst remaining competitive with SA in object discovery.

Causal discovery amounts to unearthing causal relationships amongst features in data. It is a crucial companion to causal inference, necessary to build scientific knowledge without resorting to expensive or impossible randomised control trials. In this paper, we explore how reasoning with symbolic representations can support causal discovery. Specifically, we deploy assumption-based argumentation (ABA), a well-established and powerful knowledge representation formalism, in combination with causality theories, to learn graphs which reflect causal dependencies in the data. We prove that our method exhibits desirable properties, notably that, under natural conditions, it can retrieve ground-truth causal graphs. We also conduct experiments with an implementation of our method in answer set programming (ASP) on four datasets from standard benchmarks in causal discovery, showing that our method compares well against established baselines.

There is a growing interest in understanding arguments' strength in Quantitative Bipolar Argumentation Frameworks (QBAFs). Most existing studies focus on attribution-based methods that explain an argument's strength by assigning importance scores to other arguments but fail to explain how to change the current strength to a desired one. To solve this issue, we introduce counterfactual explanations for QBAFs. We discuss problem variants and propose an iterative algorithm named Counterfactual Explanations for Quantitative bipolar Argumentation frameworks (CE-QArg). CE-QArg can identify valid and cost-effective counterfactual explanations based on two core modules, polarity and priority, which help determine the updating direction and magnitude for each argument, respectively. We discuss some formal properties of our counterfactual explanations and empirically evaluate CE-QArg on randomly generated QBAFs.

In the pursuit of enhancing the efficacy and flexibility of interpretable, data-driven classification models, this work introduces a novel incorporation of user-defined preferences with Abstract Argumentation and Case-Based Reasoning (CBR). Specifically, we introduce Preference-Based Abstract Argumentation for Case-Based Reasoning (which we call AA-CBR-P), allowing users to define multiple approaches to compare cases with an ordering that specifies their preference over these comparison approaches. We prove that the model inherently follows these preferences when making predictions and show that previous abstract argumentation for case-based reasoning approaches are insufficient at expressing preferences over constituents of an argument. We then demonstrate how this can be applied to a real-world medical dataset sourced from a clinical trial evaluating differing assessment methods of patients with a primary brain tumour. We show empirically that our approach outperforms other interpretable machine learning models on this dataset.

AI has become pervasive in recent years, but state-of-the-art approaches predominantly neglect the need for AI systems to be contestable. Instead, contestability is advocated by AI guidelines (e.g. by the OECD) and regulation of automated decision-making (e.g. GDPR). In this position paper we explore how contestability can be achieved computationally in and for AI. We argue that contestable AI requires dynamic (human-machine and/or machine-machine) explainability and decision-making processes, whereby machines can 1. interact with humans and/or other machines to progressively explain their outputs and/or their reasoning as well as assess grounds for contestation provided by these humans and/or other machines, and 2. revise their decision-making processes to redress any issues successfully raised during contestation. Given that much of the current AI landscape is tailored to static AIs, the need to accommodate contestability will require a radical rethinking, that, we argue, computational argumentation is ideally suited to support.

We present a novel framework designed to extend model reconciliation approaches, commonly used in human-aware planning, for enhanced human-AI interaction. By adopting a structured argumentation-based dialogue paradigm, our framework enables dialectical reconciliation to address knowledge discrepancies between an explainer (AI agent) and an explainee (human user), where the goal is for the explainee to understand the explainer's decision. We formally describe the operational semantics of our proposed framework, providing theoretical guarantees. We then evaluate the framework's efficacy ``in the wild'' via computational and human-subject experiments. Our findings suggest that our framework offers a promising direction for fostering effective human-AI interactions in domains where explainability is important.

Learning modular object-centric representations is said to be crucial for systematic generalization. Existing methods show promising object-binding capabilities empirically, but theoretical identifiability guarantees remain relatively underdeveloped. Understanding when object-centric representations can theoretically be identified is important for scaling slot-based methods to high-dimensional images with correctness guarantees. To that end, we propose a probabilistic slot-attention algorithm that imposes an *aggregate* mixture prior over object-centric slot representations, thereby providing slot identifiability guarantees without supervision, up to an equivalence relation. We provide empirical verification of our theoretical identifiability result using both simple 2-dimensional data and high-resolution imaging datasets.

Counterfactual Explanations (CEs) have emerged as a major paradigm in explainable AI research, providing recourse recommendations for users affected by the decisions of machine learning models. However, CEs found by existing methods often become invalid when slight changes occur in the parameters of the model they were generated for. The literature lacks a way to provide exhaustive robustness guarantees for CEs under model changes, in that existing methods to improve CEs' robustness are mostly heuristic, and the robustness performances are evaluated empirically using only a limited number of retrained models. To bridge this gap, we propose a novel interval abstraction technique for parametric machine learning models, which allows us to obtain provable robustness guarantees for CEs under a possibly infinite set of plausible model changes Δ. Based on this idea, we formalise a robustness notion for CEs, which we call Δ-robustness, in both binary and multi-class classification settings. We present procedures to verify Δ-robustness based on Mixed Integer Linear Programming, using which we further propose algorithms to generate CEs that are Δ-robust. In an extensive empirical study involving neural networks and logistic regression models, we demonstrate the practical applicability of our approach. We discuss two strategies for determining the appropriate hyperparameters in our method, and we quantitatively benchmark CEs generated by eleven methods, highlighting the effectiveness of our algorithms in finding robust CEs.

We present a principle-based analysis of contribution functions for quantitative bipolar argumentation graphs that quantify the contribution of one argument to another. The introduced principles formalise the intuitions underlying different contribution functions as well as expectations one would have regarding the behaviour of contribution functions in general. As none of the covered contribution functions satisfies all principles, our analysis can serve as a tool that enables the selection of the most suitable function based on the requirements of a given use case.

The use of counterfactual explanations (CFXs) is an increasingly popular explanation strategy for machine learning models. However, recent studies have shown that these explanations may not be robust to changes in the underlying model (e.g., following retraining), which raises questions about their reliability in real-world applications. Existing attempts towards solving this problem are heuristic, and the robustness to model changes of the resulting CFXs is evaluated with only a small number of retrained models, failing to provide exhaustive guarantees. To remedy this, we propose ∆-robustness, the first notion to formally and deterministically assess the robustness (to model changes) of CFXs for neural networks. We introduce an abstraction framework based on interval neural networks to verify the ∆-robustness of CFXs against a possibly infinite set of changes to the model parameters, i.e., weights and biases. We then demonstrate the utility of this approach in two distinct ways. First, we analyse the ∆-robustness of a number of CFX generation methods from the literature and show that they unanimously host significant deficiencies in this regard. Second, we demonstrate how embedding ∆-robustness within existing methods can provide CFXs which are provably robust.

Random forests are decision tree ensembles that can be used to solve a variety of machine learning problems. However, as the number of trees and their individual size can be large, their decision making process is often incomprehensible. We show that their decision process can be naturally represented as an argumentation problem, which allows creating global explanations via argumentative reasoning. We generalize sufficient and necessary argumentative explanations using a Markov network encoding, discuss the relevance of these explanations and establish relationships to families of abductive explanations from the literature. As the complexity of the explanation problems is high, we present an efficient approximation algorithm with probabilistic approximation guarantees.

Neural networks have proven to be effective at solving machine learning tasks but it is unclear whether they learn any relevant causal relationships, while their black-box nature makes it difficult for modellers to understand and debug them. We propose a novel method overcoming these issues by allowing a two-way interaction whereby neural-network-empowered machines can expose the underpinning learnt causal graphs and humans can contest the machines by modifying the causal graphs before re-injecting them into the machines. The learnt models are guaranteed to conform to the graphs and adhere to expert knowledge, some of which can also be given up-front. By building a window into the model behaviour and enabling knowledge injection, our method allows practitioners to debug networks based on the causal structure discovered from the data and underpinning the predictions. Experiments with real and synthetic tabular data show that our method improves predictive performance up to 2.4x while producing parsimonious networks, up to 7x smaller in the input layer, compared to SOTA regularised networks.

Argumentative explainable AI has been advocated by several in recent years, with an increasing interest on explaining the reasoning outcomes of Argumentation Frameworks (AFs). While there is a considerable body of research on qualitatively explaining the reasoning outcomes of AFs with debates/disputes/dialogues in the spirit of extension-based semantics, explaining the quantitative reasoning outcomes of AFs under gradual semantics has not received much attention, despite widespread use in applications. In this paper, we contribute to filling this gap by proposing a novel theory of Argument Attribution Explanations (AAEs) by incorporating the spirit of feature attribution from machine learning in the context of Quantitative Bipolar Argumentation Frameworks (QBAFs): whereas feature attribution is used to determine the influence of features towards outputs of machine learning models, AAEs are used to determine the influence of arguments towards topic arguments of interest. We study desirable properties of AAEs, including some new ones and some partially adapted from the literature to our setting. To demonstrate the applicability of our AAEs in practice, we conclude by carrying out two case studies in the scenarios of fake news detection and movie recommender systems.

Neural networks (NNs) have various applications in AI, but explaining their decisions remains challenging. Existing approaches often focus on explaining how changing individual inputs affects NNs’ outputs. However, an explanation that is consistent with the input-output behaviour of an NN is not necessarily faithful to the actual mechanics thereof. In this paper, we exploit relationships between multi-layer perceptrons (MLPs) and quantitative argumentation frameworks (QAFs) to create argumentative explanations for the mechanics of MLPs. Our SpArX method first sparsifies the MLP while maintaining as much of the original structure as possible. It then translates the sparse MLP into an equivalent QAF to shed light on the underlying decision process of the MLP, producing global and/or local explanations. We demonstrate experimentally that SpArX can give more faithful explanations than existing approaches, while simultaneously providing deeper insights into the actual reasoning process of MLPs.

We present LawGiBa, a proof-of-concept demonstration system for legal assistance that combines GPT, legal knowledge bases, and Prolog’s logic programming structure to provide explanations for legal queries. This novel combination effectively and feasibly addresses the hallucination issue of large language models (LLMs) in critical domains, such as law. Through this system, we demonstrate how incorporating a legal knowledge base and logical reasoning can enhance the accuracy and reliability of legal advice provided by AI models like GPT. Though our work is primarily a demonstration, it provides a framework to explore how knowledge bases and logic programming structures can be further integrated with generative AI systems, to achieve improved results across various natural languages and legal systems.

Case-based reasoning is known to play an important role in several legal settings. We focus on a recent approach to case-based reasoning, supported by an instantiation of abstract argumentation whereby arguments represent cases and attack between arguments results from outcome disagreement between cases and a notion of relevance. We explore how relevance can be learnt automatically with the help of decision trees, and explore the combination of case-based reasoning with abstract argumentation (AA-CBR) and learning of case relevance for prediction in legal settings. Specifically, we show that, for two legal datasets, AA-CBR with decision-tree-based learning of case relevance performs competitively in comparison with decision trees, and that AA-CBR with decision-tree-based learning of case relevance results in a more compact representation than their decision tree counterparts, which could facilitate cognitively tractable explanations.

As the field of explainable AI (XAI) is maturing, calls for interactive explanations for (the outputs of) AI models are growing, but the state-of-the-art predominantly focuses on static explanations. In this paper, we focus instead on interactive explanations framed as conflict resolution between agents (i.e. AI models and/or humans) by leveraging on computational argumentation. Specifically, we define Argumentative eXchanges (AXs) for dynamically sharing, in multi-agent systems, information harboured in individual agents’ quantitative bipolar argumentation frameworks towards resolving conflicts amongst the agents. We then deploy AXs in the XAI setting in which a machine and a human interact about the machine’s predictions. We identify and assess several theoretical properties characterising AXs that are suitable for XAI. Finally, we instantiate AXs for XAI by defining various agent behaviours, e.g. capturing counterfactual patterns of reasoning in machines and highlighting the effects of cognitive biases in humans. We show experimentally (in a simulated environment) the comparative advantages of these behaviours in terms of conflict resolution, and show that the strongest argument may not always be the most effective.

Deceptive agents are a challenge for the safety, trustworthiness, and cooperation of AI systems. We focus on the problem that agents might deceive in order to achieve their goals (for instance, in our experiments with language models, the goal of being evaluated as truthful). There are a number of existing definitions of deception in the literature on game theory and symbolic AI, but there is no overarching theory of deception for learning agents in games. We introduce a formal definition of deception in structural causal games, grounded in the philosophy literature, and applicable to real-world machine learning systems. Several examples and results illustrate that our formal definition aligns with the philosophical and commonsense meaning of deception. Our main technical result is to provide graphical criteria for deception. We show, experimentally, that these results can be used to mitigate deception in reinforcement learning agents and language models.

This issue of TPLP contains selected papers of the 39 th

Recent efforts within the AI community have yielded impressive results towards “soft theorem proving” over natural language sentences using language models. We propose a novel, generative adversarial framework for probing and improving these models’ reasoning capabilities. Adversarial attacks in this domain suffer from the logical inconsistency problem, whereby perturbations to the input may alter the label. Our Logically consistent AdVersarial Attacker, LAVA, addresses this by combining a structured generative process with a symbolic solver, guaranteeing logical consistency. Our framework successfully generates adversarial attacks and identifies global weaknesses common across multiple target models. Our analyses reveal naive heuristics and vulnerabilities in these models’ reasoning capabilities, exposing an incomplete grasp of logical deduction under logic programs. Finally, in addition to effective probing of these models, we show that training on the generated samples improves the target model’s performance.

Data exploration is an important step of every data science and machine learning project, including those involving textual data. We provide a novel language tool, in the form of a publicly available Python library for extracting patterns from textual data. The library integrates a first public implementation of the existing GrASP algorithm. It allows users to extract patterns using a number of general-purpose built-in linguistic attributes (such as hypernyms, part-of-speech tags, and syntactic dependency tags), as envisaged for the original algorithm, as well as domain-specific custom attributes which can be incorporated into the library by implementing two functions. The library is equipped with a web-based interface empowering human users to conveniently explore data via the extracted patterns, using complementary pattern-centric and example-centric views: the former includes a reading in natural language and statistics of each extracted pattern; the latter shows applications of each extracted pattern to training examples. We demonstrate the usefulness of the library in classification (spam detection and argument mining), model analysis (machine translation), and artifact discovery in datasets (SNLI and 20Newsgroups).

The need for manual review of various financial texts, such as company filings and news, presents a major bottleneck in financial analysts’ work. Thus, there is great potential for the application of NLP methods, tools and resources to fulfil a genuine industrial need in finance. In this paper, we show how this potential can be fulfilled by presenting an end-to-end, fully unsupervised method for knowledge discovery from financial texts. Our method creatively integrates existing resources to construct automatically a knowledge graph of companies and related entities as well as to carry out unsupervised analysis of the resulting graph to provide quantifiable and explainable insights from the produced knowledge. The graph construction integrates entity processing and semantic expansion, before carrying out open relation extraction. We illustrate our method by calculating automatically the environmental rating for companies in the S&P 500, based on company filings with the SEC (Securities and Exchange Commission). We then show the usefulness of our method in this setting by providing an assessment of our method’s outputs with an independent MSCI source.

Explainable AI (XAI) has been investigated for decades and, together with AI itself, has witnessed unprecedented growth in recent years. Among various approaches to XAI, argumentative models have been advocated in both the AI and social science literature, as their dialectical nature appears to match some basic desirable features of the explanation activity. In this survey we overview XAI approaches built using methods from the field of computational argumentation, leveraging its wide array of reasoning abstractions and explanation delivery methods. We overview the literature focusing on different types of explanation (intrinsic and post-hoc), different models with which argumentation-based explanations are deployed, different forms of delivery, and different argumentation frameworks they use. We also lay out a roadmap for future work.

Recently, abstract argumentation-based models of case-based reasoning (AA-CBR in short) have been proposed, originally inspired by the legal domain, but also applicable as classifiers in different scenarios. However, the formal properties of AA-CBR as a reasoning system remain largely unexplored. In this paper, we focus on analysing the non-monotonicity properties of a regular version of AA-CBR (that we call AA-CBR_>). Specifically, we prove that AA-CBR_> is not cautiously monotonic, a property frequently considered desirable in the literature. We then define a variation of AA-CBR_> which is cautiously monotonic. Further, we prove that such variation is equivalent to using AA-CBR_> with a restricted casebase consisting of all "surprising" and "sufficient" cases in the original casebase. As a by-product, we prove that this variation of AA-CBR_> is cumulative, rationally monotonic, and empowers a principled treatment of noise in "incoherent" casebases. Finally, we illustrate AA-CBR and cautious monotonicity questions on a case study on the U.S. Trade Secrets domain, a legal casebase.

Abstract Debugging a machine learning model is hard since the bug usually involves the training data and the learning process. This becomes even harder for an opaque deep learning model if we have no clue about how the model actually works. In this survey, we review papers that exploit explanations to enable humans to give feedback and debug NLP models. We call this problem explanation-based human debugging (EBHD). In particular, we categorize and discuss existing work along three dimensions of EBHD (the bug context, the workflow, and the experimental setting), compile findings on how EBHD components affect the feedback providers, and highlight open problems that could be future research directions.

A number of exciting advances have been made in automated fact-checking thanks to increasingly larger datasets and more powerful systems, leading to improvements in the complexity of claims which can be accurately fact-checked.However, despite these advances, there are still desirable functionalities missing from the fact-checking pipeline.In this survey, we focus on the explanation functionality -that is fact-checking systems providing reasons for their predictions.We summarize existing methods for explaining the predictions of fact-checking systems and we explore trends in this topic.Further, we consider what makes for good explanations in this specific domain through a comparative analysis of existing fact-checking explanations against some desirable properties.Finally, we propose further research directions for generating fact-checking explanations, and describe how these may lead to improvements in the research area.

Argument(ation) Mining (AM) is the research area which aims at extracting argument components and predicting argumentative relations (i.e., support and attack) from text. In particular, numerous approaches have been proposed in the literature to predict the relations holding between arguments, and application-specific annotated resources were built for this purpose. Despite the fact that these resources were created to experiment on the same task, the definition of a single relation prediction method to be successfully applied to a significant portion of these datasets is an open research problem in AM. This means that none of the methods proposed in the literature can be easily ported from one resource to another. In this paper, we address this problem by proposing a set of dataset independent strong neural baselines which obtain homogeneous results on all the datasets proposed in the literature for the argumentative relation prediction task in AM. Thus, our baselines can be employed by the AM community to compare more effectively how well a method performs on the argumentative relation prediction task.

Today’s AI landscape is permeated by plentiful data anddominated by powerful data-centric methods with the potential toimpact a wide range of human sectors. Yet, in some settings this po-tential is hindered by these data-centric AI methods being mostlyopaque. Considerable efforts are currently being devoted to defin-ing methods for explaining black-box techniques in some settings,while the use of transparent methods is being advocated in others,especially when high-stake decisions are involved, as in healthcareand the practice of law. In this paper we advocate a novel transpar-ent paradigm of Data-Empowered Argumentation (DEAr in short)for dialectically explainable predictions. DEAr relies upon the ex-traction of argumentation debates from data, so that the dialecticaloutcomes of these debates amount to predictions (e.g. classifications)that can be explained dialectically. The argumentation debates con-sist of (data) arguments which may not be linguistic in general butmay nonetheless be deemed to be ‘arguments’ in that they are dialec-tically related, for instance by disagreeing on data labels. We illus-trate and experiment with the DEAr paradigm in three settings, mak-ing use, respectively, of categorical data, (annotated) images and text.We show empirically that DEAr is competitive with another transpar-ent model, namely decision trees (DTs), while also providing natu-rally dialectical explanations.

Fact-checking is the task of verifying the veracity of claims by assessing their assertions against credible evidence.The vast majority of fact-checking studies focus exclusively on political claims.Very little research explores fact-checking for other topics, specifically subject matters for which expertise is required.We present the first study of explainable fact-checking for claims which require specific expertise.For our case study we choose the setting of public health.To support this case study we construct a new dataset PUBHEALTH of 11.8K claims accompanied by journalist crafted, gold standard explanations (i.e., judgments) to support the fact-check labels for claims 1 .We explore two tasks: veracity prediction and explanation generation.We also define and evaluate, with humans and computationally, three coherence properties of explanation quality.Our results indicate that, by training on in-domain data, gains can be made in explainable, automated fact-checking for claims which require specific expertise.

Since obtaining a perfect training dataset (i.e., a dataset which is considerably large, unbiased, and well-representative of unseen cases) is hardly possible, many real-world text classifiers are trained on the available, yet imperfect, datasets.These classifiers are thus likely to have undesirable properties.For instance, they may have biases against some sub-populations or may not work effectively in the wild due to overfitting.In this paper, we propose FINDa framework which enables humans to debug deep learning text classifiers by disabling irrelevant hidden features.Experiments show that by using FIND, humans can improve CNN text classifiers which were trained under different types of imperfect datasets (including datasets with biases and datasets with dissimilar traintest distributions).

We propose a general method for generating counterfactual explanations (CFXs) for a range of Bayesian Network Classifiers (BCs), e.g. single- or multi-label, binary or multidimensional. We focus on explanations built from relations of (critical and potential) influence between variables, indicating the reasons for classifications, rather than any probabilistic information. We show by means of a theoretical analysis of CFXs’ properties that they serve the purpose of indicating (potentially) pivotal factors in the classification process, whose absence would give rise to different classifications. We then prove empirically for various BCs that CFXs provide useful information in real world settings, e.g. when race plays a part in parole violation prediction, and show that they have inherent advantages over existing explanation methods in the literature.

As AI systems become ever more intertwined in our personal lives, the way in which they explain themselves to and interact with humans is an increasingly critical research area. The explanation of recommendations is thus a pivotal functionality in a user’s experience of a recommender system (RS), providing the possibility of enhancing many of its desirable features in addition to its effectiveness (accuracy wrt users’ preferences). For an RS that we prove empirically is effective, we show how argumentative abstractions underpinning recommendations can provide the structural scaffolding for (different types of) interactive explanations (IEs), i.e. explanations supporting interactions with users. We prove formally that these IEs empower feedback mechanisms that guarantee that recommendations will improve with time, hence rendering the RS scrutable. Finally, we prove experimentally that the various forms of IE (tabular, textual and conversational) induce trust in the recommendations and provide a high degree of transparency in the RS’s functionality.

Mathematical optimization offers highly-effective tools for finding solutions for problems with well-defined goals, notably scheduling. However, optimization solvers are often unexplainable black boxes whose solutions are inaccessible to users and which users cannot interact with. We define a novel paradigm using argumentation to empower the interaction between optimization solvers and users, supported by tractable explanations which certify or refute solutions. A solution can be from a solver or of interest to a user (in the context of ‘what-if’ scenarios). Specifically, we define argumentative and natural language explanations for why a schedule is (not) feasible, (not) efficient or (not) satisfying fixed user decisions, based on models of the fundamental makespan scheduling problem in terms of abstract argumentation frameworks (AFs). We define three types of AFs, whose stable extensions are in one-to-one correspondence with schedules that are feasible, efficient and satisfying fixed decisions, respectively. We extract the argumentative explanations from these AFs and the natural language explanations from the argumentative ones.

Piyawat Lertvittayakumjorn, Francesca Toni. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Different semantics of abstract Argumentation Frameworks (AFs) provide different levels of decisiveness for reasoning about the acceptability of conflicting arguments.The stable semantics is useful for applications requiring a high level of decisiveness, as it assigns to each argument the label "accepted" or the label "rejected". Unfortunately, stable labellings are not guaranteed to exist, thus raising the question as to which parts of AFs are responsible for the non-existence. In this paper, we address this question by investigating a more general question concerning preferred labellings (which may be less decisive than stable labellings but are always guaranteed to exist), namely why a given preferred labelling may not be stable and thus undecided on some arguments. In particular, (1) we give various characterisations of parts of an AF, based on the given preferred labelling, and (2) we show that these parts are indeed responsible for the undecisiveness if the preferred labelling is not stable. We then use these characterisations to explain the non-existence of stable labellings.

The study of properties of gradual evaluation methods in argumentation has received increasing attention in recent years, with studies devoted to various classes of frameworks/methods leading to conceptually similar but formally distinct properties in different contexts. In this paper we provide a systematic analysis for this research landscape by making three main contributions. First, we identify groups of conceptually related properties in the literature, which can be regarded as based on common patterns and, using these patterns, we evidence that many further properties can be considered. Then, we provide a simplifying and unifying perspective for these properties by showing that they are all implied by the parametric principles of (either strict or non-strict) balance and monotonicity. Finally, we show that (instances of) these principles are satisfied by several quantitative argumentation formalisms in the literature, thus confirming their general validity and their utility to support a compact, yet comprehensive, analysis of properties of gradual argumentation.

This demo presents the web system “Games of Argumentation”, which allows users to build argumentation graphs and examine them in a game-theoretical manner using up to three different evaluation techniques. The concurrent evaluations of arguments using different techniques, which may be qualitative or quantitative, provides a significant aid to users in both understanding game-theoretical argumentation semantics and pinpointing their differences from alternative semantics, traditional or otherwise, to differentiate between them.

Argumentation frameworks have been widely studied both in terms of formal properties they exhibit under different semantics and in terms of applications they can support. But where are argumentation frameworks coming from, and how can argumentation, a model-based approach to AI, beneficially integrate with the nowadays-much-widespread data-centric AI perspective? In this talk I will overview applications empowered by a variety of (extension-based and gradual) semantics for abstract and bipolar argumentation frameworks automatically obtained from data (including but not limited to text) and from logical rules. Some of these applications require the integration of argumentation and machine learning, and result in a mixed model-based and data-centric pipeline. For some applications, the semantics informs the definition of the frameworks rather than, as is conventionally the case, being enforced on frameworks a posteriori.

A significant problem of recommender systems is their inability to explain recommendations, resulting in turn in ineffective feedback from users and the inability to adapt to users’ preferences. We propose a hybrid method for calculating predicted ratings, built upon an item/aspect-based graph with users’ partially given ratings, that can be naturally used to provide explanations for recommendations, extracted from user-tailored Tripolar Argumentation Frameworks (TFs). We show that our method can be understood as a gradual semantics for TFs, exhibiting a desirable, albeit weak, property of balance. We also show experimentally that our method is competitive in generating correct predictions, compared with state-of-the-art methods, and illustrate how users can interact with the generated explanations to improve quality of recommendations.

Different semantics of abstract Argumentation Frameworks (AFs) provide different levels of decisiveness for reasoning about the acceptability of conflicting arguments. The stable semantics is useful for applications requiring a high level of decisiveness, as it assigns to each argument the label “accepted” or the label “rejected”. Unfortunately, stable labellings are not guaranteed to exist, thus raising the question as to which parts of AFs are responsible for the non-existence. In this paper, we address this question by investigating a more general question concerning preferred labellings (which may be less decisive than stable labellings but are always guaranteed to exist), namely why a given preferred labelling may not be stable and thus undecided on some arguments. In particular, (1) we give various characterisations of parts of an AF, based on the given preferred labelling, and (2) we show that these parts are indeed responsible for the undecisiveness if the preferred labelling is not stable. We then use these characterisations to explain the non-existence of stable labellings. We present two types of characterisations, based on labellings that are more (or equally) committed than the given preferred labelling on the one hand, and based on the structure of the given AF on the other, and compare the respective AF parts deemed responsible. To prove that our characterisations indeed yield responsible parts, we use a notion of enforcement of labels through structural revision, by means of which the preferred labelling of the given AF can be turned into a stable labelling of the structurally revised AF. Rather than prescribing how this structural revision is carried out, we focus on the enforcement of labels and leave the engineering of the revision open to fulfil differing requirements of applications and information available to users.

The use of social media has become a regular habit for many and has changed the way people interact with each other. In this article, we focus on analyzing whether news headlines support tweets and whether reviews are deceptive by analyzing the interaction or the influence that these texts have on the others, thus exploiting contextual information. Concretely, we define a deep learning method for relation–based argument mining to extract argumentative relations of attack and support. We then use this method for determining whether news articles support tweets, a useful task in fact-checking settings, where determining agreement toward a statement is a useful step toward determining its truthfulness. Furthermore, we use our method for extracting bipolar argumentation frameworks from reviews to help detect whether they are deceptive. We show experimentally that our method performs well in both settings. In particular, in the case of deception detection, our method contributes a novel argumentative feature that, when used in combination with other features in standard supervised classifiers, outperforms the latter even on small data sets.

We propose a deep learning architecture to capture argumentative relations of attack and support from one piece of text to another, of the kind that naturally occur in a debate.The architecture uses two (unidirectional or bidirectional) Long Short-Term Memory networks and (trained or non-trained) word embeddings, and allows to considerably improve upon existing techniques that use syntactic features and supervised classifiers for the same form of (relation-based) argument mining.

Argumentation has proven successful in a number of domains, including Multi-Agent Systems and decision support in medicine and engineering. We propose its application to a domain yet largely unexplored by argumentation research: computational linguistics. We have developed a novel classification methodology that incorporates reasoning through argumentation with supervised learning. We train classifiers and then argue about the validity of their output. To do so, we identify arguments that formalise prototypical knowledge of a problem and use them to correct misclassifications. We illustrate our methodology on two tasks. On the one hand, we address cross-domain sentiment polarity classification , where we train classifiers on one corpus, for example, Tweets, to identify positive/negative polarity and classify instances from another corpus, for example, sentences from movie reviews. On the other hand, we address a form of argumentation mining that we call Relation-based Argumentation Mining , where we classify pairs of sentences based on whether the first sentence attacks or supports the second or whether it does neither. Whenever we find that one sentence attacks/supports the other, we consider both to be argumentative, irrespective of their stand-alone argumentativeness. For both tasks, we improve classification performance when using our methodology, compared to using standard classifiers only.

Existing approaches using argumentation to aid or improve machine learning differ in the type of machine learning technique they consider, in their use of argumentation and in their choice of argumentation framework and semantics. This paper presents a survey of this relatively young field highlighting, in particular, its achievements to date, the applications it has been used for as well as the benefits brought about by the use of argumentation, with an eye towards its future.

The unstoppable rise of social networks and the web is facing a serious challenge: identifying the truthfulness of online opinions and reviews. We propose a system to identify two new argumentative features that a trained classifier can use to help determine whether a review is deceptive.

Case-based reasoning (CBR) is extensively used in AI in support of several applications, to assess a new situation (or case) by recollecting past situations (or cases) and employing the ones most similar to the new situation to give the assessment. In this paper we study properties of a recently proposed method for CBR, based on instantiated Abstract Argumentation and referred to as AA-CBR, for problems where cases are represented by abstract factors and (positive or negative) outcomes, and an outcome for a new case, represented by abstract factors, needs to be established. In addition, we study properties of explanations in AA-CBR and define a new notion of lean explanations that utilize solely relevant cases. Both forms of explanations can be seen as dialogical processes between a proponent and an opponent, with the burden of proof falling on the proponent.

Abstract An answer set is a plain set of literals which has no further structure that would explain why certain literals are part of it and why others are not. We show how argumentation theory can help to explain why a literal is or is not contained in a given answer set by defining two justification methods, both of which make use of the correspondence between answer sets of a logic program and stable extensions of the assumption-based argumentation (ABA) framework constructed from the same logic program.Attack Treesjustify a literal in argumentation-theoretic terms, i.e. using arguments and attacks between them, whereasABA-Based Answer Set Justificationsexpress the same justification structure in logic programming terms, that is using literals and their relationships. Interestingly, an ABA-Based Answer Set Justification corresponds to an admissible fragment of the answer set in question, and an Attack Tree corresponds to an admissible fragment of the stable extension corresponding to this answer set.

Logic Programming and Argumentation Theory have been existing side by side as two separate, yet related, techniques in the field of Knowledge Representation and Reasoningfor many years.When Assumption-Based Argumentation (ABA) was first introduced in the nineties,the authors showed how a logic program can be encoded in an ABA framework andproved that the stable semantics of a logic program corresponds to the stable extension semantics of the ABA framework encoding this logic program.We revisit this initial work by provingthat the 3-valued stable semantics of a logic program coincides with the complete semantics of the encoding ABA framework,and that the L-stable semantics of this logic program coincides with the semi-stable semantics of the encoding ABA framework.Furthermore, we show how to graphically represent the structure of a logic program encoded in an ABA frameworkand that not only logic programming and ABA semanticsbut also Abstract Argumentation semantics can be easily applied to a logic program using these graphical representations.

Argumentation can be viewed as a process of generating explanations. However, existing argumentation semantics are developed for identifying acceptable arguments within a set, rather than giving concrete justifications for them. In this work, we propose a new argumentation semantics, related admissibility, designed for giving explanations to arguments in both Abstract Argumentation and Assumption-based Argumentation. We identify different types of explanations defined in terms of the new semantics. We also give a correct computational counterpart for explanations using dispute forests.

The 31st edition of the International Conference of Logic Programming (ICLP 2015) took place in Cork, Ireland, from 31 August 2015 to 4 September 2015, co-located with the 21st International Conference on Principles and Practice of Constraint Programming (CP 2015) and part of George Boole 200, a celebration of the life and work of George Boole who was born in 1815 and worked at the University College of Cork.

Recently, argument labellings have been proposed as a new (equivalent) way to express the extension semantics of Abstract Argumentation (AA) frame-works. Here, we introduce a labelling approach for the complete semantics in Assumption-Based Argumentation (ABA), where labels are assigned to assumptions rather than whole arguments. We prove that the complete assumption labelling corresponds to the complete extension semantics in ABA, as well as to the complete extension semantics and the complete argument labelling in AA.

With this demonstration we introduce AFAlpha, a prototype of an Argument Mining tool, working in unison with Quaestio-it, an online social debating platform. While AFAlpha extracts arguments from text, as well as attack and support relations between arguments, Quaestio-it is concerned with visualising and evaluating interactive debates. We thus use Quaestio-it to represent and visualise output from AFAlpha, with the goal of taking plain text, in our case online reviews, and representing it in the form of a debate.

We propose a novel logic-based argumentation framework, called Argumentation Logic (AL), built upon a restriction of classical Propositional Logic (PL) as its underlying logic. This allows us to control the application of Reduction ad Absurdum (RA). In the case of classically consistent theories, AL and PL are equivalent, and RA is recovered through a notion of (non-)acceptability of arguments. In the case of classically inconsistent theories, AL is an extension of PL that does not trivialize, enjoying good logic-based argumentation and general logical properties.

Argumentation can be viewed as a process of generating explanations. We propose a new argumentation semantics, related admissibility, for closely capturing explanations in Abstract Argumentation, and distinguish between compact and verbose explanations. We show that dispute forests, composed of dispute trees, can be used to correctly compute these explanations.

Multi-Agent Learning is a complex problem, especially in real-time systems. We address this problem by introducing Argumentation Accelerated Reinforcement Learning (AARL), which provides a methodology for defining heuristics, represented by arguments, and incorporates these heuristics into Reinforcement Learning (RL) by using reward shaping. We define AARL via argumentation and prove that it can coordinate independent cooperative agents that have a shared goal but need to perform different actions. We test AARL empirically in a popular RL testbed, RoboCup Takeaway, and show that it significantly improves upon standard RL.

We present a system for collaborative decision support in design based upon argumentation techniques integrated with two additional AI techniques (case-based reasoning and reasoning with ontologies) so as to benefit one another and provide enriched functionalities. We evaluate these functionalities in the context of engineering design problems in injection moulding, in the context of the EU Des-MOLD project.

We present a method for (two) agents to resolve conflicts amongst themselves, when these conflicts arise from the agents suggesting different realizations of the same goal. The method uses generalpurpose dialogues to allow agents to exchange views. These are in the form of rules, assumptions and contraries of assumptions, in the format underlying Assumption-Based Argumentation (ABA). Thus, the dialogues amount to conducting an argumentation process in ABA. We define successful dialogues as those giving admissible sets of arguments and prove that these successful dialogues resolve conflicts. Thus, we provide a fully distributed, argumentation-based solution to conflict resolution while at the same time linking the computation of a well-known argumentation semantics and two-agent conflict resolution.

Recently we have seen a few development in argumentation-based dialogue systems, but there is less research in understanding agents' strategic behaviour in dialogues. We study agent strategies by linking a specific form of argumentation-based dialogues and mechanism design. Specifically, focusing on persuasion dialogues, we show how dialogues can be mapped to concepts in mechanism design. We prove that a “truthful” and “thorough” dialogue strategy is a dominant strategy under specific conditions. We also prove that a mechanism using this dialogue strategy implements a “persuasion social choice function” we define. These results show the validity of the proposed strategies for agents in persuasion and the feasibility of studying persuasion with mechanism design techniques.

Reinforcement Learning (RL) suffers from several difficulties when applied to domains with no obvious goal-state defined; this leads to inefficiency in RL algorithms. We consider a solution within the context of a widely-used testbed for RL: RoboCup Keepaway. We introduce Argumentation-Based RL (ABRL), using methods from argumentation theory to integrate domain knowledge, represented by arguments, into the SMDP algorithm for RL by using potential-based reward shaping. Empirical results show that ABRL outperforms the original SMDP algorithm, for this game, by improving convergence speed and optimality.

S. Modgil, F. Toni; Special Issue on Argumentation in Agreement Technologies, Journal of Logic and Computation, Volume 22, Issue 5, 1 October 2012, Pages 9

Formal argumentation based dialogue models have attracted some research interests recently. Within this line of research, we propose a formal model for argumentation-based dialogues between agents, using assumption-based argumentation (ABA). Thus, the dialogues amount to conducting an argumentation process in ABA. The model is given in terms of ABA-specific utterances, debate trees and forests implicitly built during and drawn from dialogues, legal-move functions (amounting to protocols) and outcome functions. Moreover, we investigate the strategic behaviour of agents in dialogues, using strategy-move functions. We instantiate our dialogue model in a range of dialogue types studied in the literature, including information-seeking, inquiry, persuasion, conflict resolution, and discovery. Finally, we prove (1) a formal connection between dialogues and well-known argumentation semantics, and (2) soundness and completeness results for our dialogue models and dialogue strategies used in different dialogue types.

We give some design guidelines for argumentation systems. These guidelines are meant to indicate essential features of argumentation when used to support “practical reasoning”. We express the guidelines in terms of postulates. We use a notion of redundancy to provide a formal counterpart of these postulates. We study the satisfaction of these postulates in two existing argumentation frameworks: assumption-based argumentation and argumentation in classical logic.

Conflicts exist in multi-agent systems. Agents have different interests and desires. Agents also hold different beliefs and may make different assumptions. To resolve conflicts, agents need to better convey information between each other and facilitate fair negotiations that yield jointly agreeable outcomes. In this paper, we present a two-agent conflict resolution scheme developed under Assumption-Based Argumentation (ABA). Agents represent their beliefs and desires in ABA. Conflicts are resolved by merging conflicting arguments. We also discuss the notion of fairness and the use of argumentation dialogue in conflict resolution.

Abstract We present the CIFF proof procedure for abductive logic programming with constraints, and we prove its correctness. CIFF is an extension of the IFF proof procedure for abductive logic programming, relaxing the original restrictions over variable quantification (allowedness conditions) and incorporating a constraint solver to deal with numerical constraints as in constraint logic programming. Finally, we describe the CIFF system, comparing it with state-of-the-art abductive systems and answer set solvers and showing how to use it to program some applications.

This paper presents the computational logic foundations of a model of agency called the KGP (Knowledge, Goals and Plan model. This model allows the specification of heterogeneous agents that can interact with each other, and can exhibit both proactive and reactive behaviour allowing them to function in dynamic environments by adjusting their goals and plans when changes happen in such environments. KGP provides a highly modular agent architecture that integrates a collection of reasoning and physical capabilities, synthesised within transitions that update the agent's state in response to reasoning, sensing and acting. Transitions are orchestrated by cycle theories that specify the order in which transitions are executed while taking into account the dynamic context and agent preferences, as well as selection operators for providing inputs to transitions.

A multiagent system may be thought of as an artificial society of autonomous software agents and we can apply concepts borrowed from welfare economics and social choice theory to assess the social welfare of such an agent society. In this paper, we study an abstract negotiation framework where agents can agree on multilateral deals to exchange bundles of indivisible resources. We then analyse how these deals affect social welfare for different instances of the basic framework and different interpretations of the concept of social welfare itself. In particular, we show how certain classes of deals are both sufficient and necessary to guarantee that a socially optimal allocation of resources will be reached eventually.

In recent years, argumentation has been shown to be an appropriate framework in which logic programming with negation as failure as well as other logics for non-monotonic reasoning can be encompassed. Many of the existing semantics for negation as failure in logic programming can be understood in a uniform way using argumentation. Moreover, other logics for non-monotonic reasoning that can also be formulated via argumentation can be given new semantics, by a direct extension of the logic programming semantics. In this paper we develop an abstract computational framework where various argumentation semantics can be computed via different parametric variations of a simple basic proof theory. This proof theory is given in terms of derivations of trees where each node in a tree contains an argument (or attack) against its corresponding parent node. The proposed proof theory, defined here for the case of logic programming, generalizes directly to other logics for non-monotonic reasoning that can also be formalized via argumentation. The abstract proof theory forms the basis for developing concrete top-down proof procedures for query evaluation. These proof procedures are obtained by adopting specific search strategies and ways of computing attacks in the particular argumentation framework. For logic programming these procedures can be seen as a generalization of the Eshghi-Kowalski abductive proof procedure that in turn generalizes SLDNF.

We present an extension of Logic Programming (LP) which, in addition to ordinary LP clauses, also includes integrity constraints, explicit representation of disjunction in the bodies of clauses and in goals, and suspension of atoms as in concurrent l

This paper is a survey and critical overview of recent work on the extension of logic programming to perform abductive reasoning (abductive logic programming). We outline the general framework of abduction and its applications to knowledge assimilation and default reasoning; and we introduce an argumentation-theoretic approach to the use of abduction as an interpretation for negation as failure. We also analyse the links between abduction and the extension of logic programming obtained by adding a form of explicit negation. Finally we discuss the relation between abduction and truth maintenance.