VULCA-Bench: A Multicultural Vision-Language Benchmark for Evaluating Cultural Understanding

We introduce VULCA-Bench, a multicultural art-critique benchmark for evaluating Vision-Language Models' (VLMs) cultural understanding beyond surface-level visual perception. Existing VLM benchmarks predominantly measure L1-L2 capabilities (object recognition, scene description, and factual question answering) while under-evaluate higher-order cultural interpretation. VULCA-Bench contains 7,410 matched image-critique pairs spanning eight cultural traditions, with Chinese-English bilingual coverage. We operationalise cultural understanding using a five-layer framework (L1-L5, from Visual Perception to Philosophical Aesthetics), instantiated as 225 culture-specific dimensions and supported by expert-written bilingual critiques. Our pilot results indicate that higher-layer reasoning (L3-L5) is consistently more challenging than visual and technical analysis (L1-L2). The dataset, evaluation scripts, and annotation tools are available under CC BY 4.0 in the supplementary materials.

Cross-Cultural Expert-Level Art Critique Evaluation with Vision-Language Models

Vision-Language Models (VLMs) excel at visual perception, yet their ability to interpret cultural meaning in art remains under-validated. We present a tri-tier evaluation framework for cross-cultural art-critique assessment: Tier I computes automated coverage and risk indicators offline; Tier II applies rubric-based scoring using a single primary judge across five dimensions; and Tier III calibrates the Tier II aggregate score to human ratings via isotonic regression, yielding a 5.2% reduction in MAE on a 152-sample held-out set. The framework outputs a calibrated cultural-understanding score for model selection and cultural-gap diagnosis, together with dimension-level diagnostics and risk indicators. We evaluate 15 VLMs on 294 expert anchors spanning six cultural traditions. Key findings are that (i) automated metrics are unreliable proxies for cultural depth, (ii) Western samples score higher than non-Western samples under our sampling and rubric, and (iii) cross-judge scale mismatch makes naive score averaging unreliable, motivating a single primary judge with explicit calibration. Dataset and code are available in the supplementary materials.

An Explainable Framework for Misinformation Identification via Critical Question Answering

Natural language misinformation detection approaches have been, to date, largely dependent on sequence classification methods, producing opaque systems in which the reasons behind classification as misinformation are unclear. While an effort has been made in the area of automated fact-checking to propose explainable approaches to the problem, this is not the case for automated reason-checking systems. In this paper, we propose a new explainable framework for both factual and rational misinformation detection based on the theory of Argumentation Schemes and Critical Questions. For that purpose, we create and release NLAS-CQ, the first corpus combining 3,566 textbook-like natural language argumentation scheme instances and 4,687 corresponding answers to critical questions related to these arguments. On the basis of this corpus, we implement and validate our new framework which combines classification with question answering to analyse arguments in search of misinformation, and provides the explanations in form of critical questions to the human user.

NLAS-multi: A Multilingual Corpus of Automatically Generated Natural Language Argumentation Schemes

Some of the major limitations identified in the areas of argument mining, argument generation, and natural language argument analysis are related to the complexity of annotating argumentatively rich data, the limited size of these corpora, and the constraints that represent the different languages and domains in which these data is annotated. To address these limitations, in this paper we present the following contributions: (i) an effective methodology for the automatic generation of natural language arguments in different topics and languages, (ii) the largest publicly available corpus of natural language argumentation schemes, and (iii) a set of solid baselines and fine-tuned models for the automatic identification of argumentation schemes.

VivesDebate-Speech: A Corpus of Spoken Argumentation to Leverage Audio Features for Argument Mining

In this paper, we describe VivesDebate-Speech, a corpus of spoken argumentation created to leverage audio features for argument mining tasks. The creation of this corpus represents an important contribution to the intersection of speech processing and argument mining communities, and one of the most complete publicly available resources in this topic. Moreover, we have performed a set of first-of-their-kind experiments which show an improvement when integrating audio features into the argument mining pipeline. The provided results can be used as a baseline for future research.

A Cascade Model for Argument Mining in Japanese Political Discussions: the QA Lab-PoliInfo-3 Case Study

The rVRAIN team tackled the Budget Argument Mining (BAM) task, consisting of a combination of classification and information retrieval sub-tasks. For the argument classification (AC), the team achieved its best performing results with a five-class BERT-based cascade model complemented with some handcrafted rules. The rules were used to determine if the expression was monetary or not. Then, each monetary expression was classified as a premise or as a conclusion in the first level of the cascade model. Finally, each premise was classified into the three premise classes, and each conclusion into the two conclusion classes. For the information retrieval (i.e., relation ID detection or RID), our best results were achieved by a combination of a BERT-based binary classifier, and the cosine similarity of pairs consisting of the monetary expression and budget dense embeddings.

Automatic Debate Evaluation with Argumentation Semantics and Natural Language Argument Graph Networks

The lack of annotated data on professional argumentation and complete argumentative debates has led to the oversimplification and the inability of approaching more complex natural language processing tasks. Such is the case of the automatic debate evaluation. In this paper, we propose an original hybrid method to automatically evaluate argumentative debates. For that purpose, we combine concepts from argumentation theory such as argumentation frameworks and semantics, with Transformer-based architectures and neural graph networks. Furthermore, we obtain promising results that lay the basis on an unexplored new instance of the automatic analysis of natural language arguments.

Transformer-Based Models for Automatic Identification of Argument Relations: A Cross-Domain Evaluation

Argument Mining is defined as the task of automatically identifying and extracting argumentative components (e.g., premises, claims, etc.) and detecting the existing relations among them (i.e., support, attack, rephrase, no relation). One of the main issues when approaching this problem is the lack of data, and the size of the publicly available corpora. In this work, we use the recently annotated US2016 debate corpus. US2016 is the largest existing argument annotated corpus, which allows exploring the benefits of the most recent advances in Natural Language Processing in a complex domain like Argument (relation) Mining. We present an exhaustive analysis of the behavior of transformer-based models (i.e., BERT, XLNET, RoBERTa, DistilBERT and ALBERT) when predicting argument relations. Finally, we evaluate the models in five different domains, with the objective of finding the less domain dependent model. We obtain a macro F1-score of 0.70 with the US2016 evaluation corpus, and a macro F1-score of 0.61 with the Moral Maze cross-domain corpus.