Konstantinos Bougiatiotis

Dimitrios Kelesis, Konstantinos Bougiatiotis, Georgios Paliouras

Although large language models (LLMs) have set benchmarks for zero-shot reasoning, their deployment remains cost-prohibitive and environmentally taxing. Small Language Models (SLMs) offer a sustainable alternative, but prone to errors, on tasks requiring complex, multi-hop logical grounding. We investigate a neuro-symbolic agentic framework to enhance the reasoning capabilities of SLMs, specifically Gemma 3 (1B, 4B) and Llama 3.2 (3B), using the CLUTRR kinship benchmark. Our approach transforms the SLM into a minimalist agent utilizing two specialized tool calls: extract_facts for symbolic triplet extraction and get_hint for expert reasoning via a Relational Graph Convolutional Network (RGCN). We evaluate these models across two configurations, both in an Oracle scenario with ground-truth triplets and a Realistic scenario relying on self-extracted knowledge. Our results reveal that while RGCN-derived hints provide a 1.5 - 2x performance gain over story-only baselines, the system is constrained by the extraction bottleneck and sequential deductive fragility, where early extraction errors compound over multi-hop chains. Furthermore, we identify a "distraction effect" in specific architectures where noisy, self-generated facts degrade performance despite the presence of expert hints. This work characterizes the challenges of symbolic grounding in low-resource agentic systems and provides a roadmap for iterative verification in neuro-symbolic agentic pipelines.

Konstantinos Bougiatiotis, Dimitrios Kelesis, Georgios Paliouras

Small language models (SLMs) have shown promise for zero-shot molecular property prediction from SMILES strings, yet they often suffer from structural blindness because sequence representations under-specify key graph-topological cues. We propose a modular Context-Augmented Prompting framework that enables agentic tool use at inference time: a trained GNN expert model provides a predictive hint with confidence, and a GNN extracts an instance-specific explanatory subgraph (e.g., a subgraph SMILES and an accompanying explanatory paragraph). We evaluate three commonly used SLMs on MUTAG and Tox21 under five prompting configurations ranging from SMILES-only to using all available tools at hand. Across two datasets, enriching prompts with graph-derived context yields substantial accuracy gains, often exceeding 25% relative improvement and up to 74% on Tox21. We further validate the functional relevance of the extracted motifs via a necessity-based edge-drop intervention. Despite the observed gains, a persistent gap remains to specialized GNN models, highlighting both the value and limits of text-conditioned reasoning for molecular structure.

Konstantinos Bougiatiotis, Georgios Paliouras

In this work, we propose a novel representation of complex multi-relational networks, which is compact and allows very efficient network analysis. Multi-relational networks capture complex data relationships and have a variety of applications, ranging from biomedical to financial, social, etc. As they get to be used with ever larger quantities of data, it is crucial to find efficient ways to represent and analyse such networks. This paper introduces the concept of Prime Adjacency Matrices (PAMs), which utilize prime numbers, to represent the relations of the network. Due to the fundamental theorem of arithmetic, this allows for a lossless, compact representation of a complete multi-relational graph, using a single adjacency matrix. Moreover, this representation enables the fast computation of multi-hop adjacency matrices, which can be useful for a variety of downstream tasks. We illustrate the benefits of using the proposed approach through various simple and complex network analysis tasks.

Anastasios Nentidis, Konstantinos Bougiatiotis, Anastasia Krithara et al.

In biomedical research, unified access to up-to-date domain-specific knowledge is crucial, as such knowledge is continuously accumulated in scientific literature and structured resources. Identifying and extracting specific information is a challenging task and computational analysis of knowledge bases can be valuable in this direction. However, for disease-specific analyses researchers often need to compile their own datasets, integrating knowledge from different resources, or reuse existing datasets, that can be out-of-date. In this study, we propose a framework to automatically retrieve and integrate disease-specific knowledge into an up-to-date semantic graph, the iASiS Open Data Graph. This disease-specific semantic graph provides access to knowledge relevant to specific concepts and their individual aspects, in the form of concept relations and attributes. The proposed approach is implemented as an open-source framework and applied to three diseases (Lung Cancer, Dementia, and Duchenne Muscular Dystrophy). Exemplary queries are presented, investigating the potential of this automatically generated semantic graph as a basis for retrieval and analysis of disease-specific knowledge.

Konstantinos Bougiatiotis, Georgios Paliouras

Multi-relational networks capture intricate relationships in data and have diverse applications across fields such as biomedical, financial, and social sciences. As networks derived from increasingly large datasets become more common, identifying efficient methods for representing and analyzing them becomes crucial. This work extends the Prime Adjacency Matrices (PAMs) framework, which employs prime numbers to represent distinct relations within a network uniquely. This enables a compact representation of a complete multi-relational graph using a single adjacency matrix, which, in turn, facilitates quick computation of multi-hop adjacency matrices. In this work, we enhance the framework by introducing a lossless algorithm for calculating the multi-hop matrices and propose the Bag of Paths (BoP) representation, a versatile feature extraction methodology for various graph analytics tasks, at the node, edge, and graph level. We demonstrate the efficiency of the framework across various tasks and datasets, showing that simple BoP-based models perform comparably to or better than commonly used neural models while offering improved speed and interpretability.

Elias Zavitsanos, Dimitris Mavroeidis, Konstantinos Bougiatiotis et al.

In this work, we examine the evaluation process for the task of detecting financial reports with a high risk of containing a misstatement. This task is often referred to, in the literature, as ``misstatement detection in financial reports''. We provide an extensive review of the related literature. We propose a new, realistic evaluation framework for the task which, unlike a large part of the previous work: (a) focuses on the misstatement class and its rarity, (b) considers the dimension of time when splitting data into training and test and (c) considers the fact that misstatements can take a long time to detect. Most importantly, we show that the evaluation process significantly affects system performance, and we analyze the performance of different models and feature types in the new realistic framework.

Konstantinos Bougiatiotis, Georgios Paliouras

Most phenomena related to biomedical tasks are inherently complex, and in many cases, are expressed as signals on biomedical Knowledge Graphs (KGs). In this work, we introduce the use of a new representation framework, the Prime Adjacency Matrix (PAM) for biomedical KGs, which allows for very efficient network analysis. PAM utilizes prime numbers to enable representing the whole KG with a single adjacency matrix and the fast computation of multiple properties of the network. We illustrate the applicability of the framework in the biomedical domain by working on different biomedical knowledge graphs and by providing two case studies: one on drug-repurposing for COVID-19 and one on important metapath extraction. We show that we achieve better results than the original proposed workflows, using very simple methods that require no training, in considerably less time.

Lefteris Loukas, Konstantinos Bougiatiotis, Manos Fergadiotis et al.

We present the submission of team DICoE for FinSim-3, the 3rd Shared Task on Learning Semantic Similarities for the Financial Domain. The task provides a set of terms in the financial domain and requires to classify them into the most relevant hypernym from a financial ontology. After augmenting the terms with their Investopedia definitions, our system employs a Logistic Regression classifier over financial word embeddings and a mix of hand-crafted and distance-based features. Also, for the first time in this task, we employ different replacement methods for out-of-vocabulary terms, leading to improved performance. Finally, we have also experimented with word representations generated from various financial corpora. Our best-performing submission ranked 4th on the task's leaderboard.

Anastasios Nentidis, Anastasia Krithara, Konstantinos Bougiatiotis et al.

In this paper, we present an overview of the eighth edition of the BioASQ challenge, which ran as a lab in the Conference and Labs of the Evaluation Forum (CLEF) 2020. BioASQ is a series of challenges aiming at the promotion of systems and methodologies for large-scale biomedical semantic indexing and question answering. To this end, shared tasks are organized yearly since 2012, where different teams develop systems that compete on the same demanding benchmark datasets that represent the real information needs of experts in the biomedical domain. This year, the challenge has been extended with the introduction of a new task on medical semantic indexing in Spanish. In total, 34 teams with more than 100 systems participated in the three tasks of the challenge. As in previous years, the results of the evaluation reveal that the top-performing systems managed to outperform the strong baselines, which suggests that state-of-the-art systems keep pushing the frontier of research through continuous improvements.

K. Bougiatiotis, R. Fasoulis, F. Aisopos et al.

Nowadays Knowledge Graphs constitute a mainstream approach for the representation of relational information on big heterogeneous data, however, they may contain a big amount of imputed noise when constructed automatically. To address this problem, different error detection methodologies have been proposed, mainly focusing on path ranking and representation learning. This work presents various mainstream approaches and proposes a hybrid and modular methodology for the task. We compare different methods on two benchmarks and one real-world biomedical publications dataset, showcasing the potential of our approach and providing insights on graph embeddings when dealing with noisy Knowledge Graphs.

Konstantinos Bougiatiotis, Theodore Giannakopoulos

In this paper we examine the ability of low-level multimodal features to extract movie similarity, in the context of a content-based movie recommendation approach. In particular, we demonstrate the extraction of multimodal representation models of movies, based on textual information from subtitles, as well as cues from the audio and visual channels. With regards to the textual domain, we emphasize our research in topic modeling of movies based on their subtitles, in order to extract topics that discriminate between movies. Regarding the visual domain, we focus on the extraction of semantically useful features that model camera movements, colors and faces, while for the audio domain we adopt simple classification aggregates based on pretrained models. The three domains are combined with static metadata (e.g. directors, actors) to prove that the content-based movie similarity procedure can be enhanced with low-level multimodal information. In order to demonstrate the proposed content representation approach, we have built a small dataset of 160 widely known movies. We assert movie similarities, as propagated by the individual modalities and fusion models, in the form of recommendation rankings. Extensive experimentation proves that all three low-level modalities (text, audio and visual) boost the performance of a content-based recommendation system, compared to the typical metadata-based content representation, by more than $50\%$ relative increase. To our knowledge, this is the first approach that utilizes a wide range of features from all involved modalities, in order to enhance the performance of the content similarity estimation, compared to the metadata-based approaches.

Konstantinos Bougiatiotis, Theodore Giannakopoulos

In this paper we examine the existence of correlation between movie similarity and low level features from respective movie content. In particular, we demonstrate the extraction of multi-modal representation models of movies based on subtitles, audio and metadata mining. We emphasize our research in topic modeling of movies based on their subtitles. In order to demonstrate the proposed content representation approach, we have built a small dataset of 160 widely known movies. We assert movie similarities, as propagated by the singular modalities and fusion models, in the form of recommendation rankings. We showcase a novel topic model browser for movies that allows for exploration of the different aspects of similarities between movies and an information retrieval system for movie similarity based on multi-modal content.