Sheeba Samuel

Jihen Amara, Birgitta König-Ries, Sheeba Samuel

Recently, there has been a growing interest in Multimodal Large Language Models (MLLMs) due to their remarkable potential in various tasks integrating different modalities, such as image and text, as well as applications such as image captioning and visual question answering. However, such models still face challenges in accurately captioning and interpreting specific visual concepts and classes, particularly in domain-specific applications. We argue that integrating domain knowledge in the form of an ontology can significantly address these issues. In this work, as a proof of concept, we propose a new framework that combines ontology with MLLMs to classify images of plant diseases. Our method uses concepts about plant diseases from an existing disease ontology to query MLLMs and extract relevant visual concepts from images. Then, we use the reasoning capabilities of the ontology to classify the disease according to the identified concepts. Ensuring that the model accurately uses the concepts describing the disease is crucial in domain-specific applications. By employing an ontology, we can assist in verifying this alignment. Additionally, using the ontology's inference capabilities increases transparency, explainability, and trust in the decision-making process while serving as a judge by checking if the annotations of the concepts by MLLMs are aligned with those in the ontology and displaying the rationales behind their errors. Our framework offers a new direction for synergizing ontologies and MLLMs, supported by an empirical study using different well-known MLLMs.

Jihen Amara, Birgitta König-Ries, Sheeba Samuel

Plant diseases remain a considerable threat to food security and agricultural sustainability. Rapid and early identification of these diseases has become a significant concern motivating several studies to rely on the increasing global digitalization and the recent advances in computer vision based on deep learning. In fact, plant disease classification based on deep convolutional neural networks has shown impressive performance. However, these methods have yet to be adopted globally due to concerns regarding their robustness, transparency, and the lack of explainability compared with their human experts counterparts. Methods such as saliency-based approaches associating the network output to perturbations of the input pixels have been proposed to give insights into these algorithms. Still, they are not easily comprehensible and not intuitive for human users and are threatened by bias. In this work, we deploy a method called Testing with Concept Activation Vectors (TCAV) that shifts the focus from pixels to user-defined concepts. To the best of our knowledge, our paper is the first to employ this method in the field of plant disease classification. Important concepts such as color, texture and disease related concepts were analyzed. The results suggest that concept-based explanation methods can significantly benefit automated plant disease identification.

Vamsi Krishna Kommineni, Birgitta König-Ries, Sheeba Samuel

The conventional process of building Ontologies and Knowledge Graphs (KGs) heavily relies on human domain experts to define entities and relationship types, establish hierarchies, maintain relevance to the domain, fill the ABox (or populate with instances), and ensure data quality (including amongst others accuracy and completeness). On the other hand, Large Language Models (LLMs) have recently gained popularity for their ability to understand and generate human-like natural language, offering promising ways to automate aspects of this process. This work explores the (semi-)automatic construction of KGs facilitated by open-source LLMs. Our pipeline involves formulating competency questions (CQs), developing an ontology (TBox) based on these CQs, constructing KGs using the developed ontology, and evaluating the resultant KG with minimal to no involvement of human experts. We showcase the feasibility of our semi-automated pipeline by creating a KG on deep learning methodologies by exploiting scholarly publications. To evaluate the answers generated via Retrieval-Augmented-Generation (RAG) as well as the KG concepts automatically extracted using LLMs, we design a judge LLM, which rates the generated content based on ground truth. Our findings suggest that employing LLMs could potentially reduce the human effort involved in the construction of KGs, although a human-in-the-loop approach is recommended to evaluate automatically generated KGs.

Waqas Ahmed, Sheeba Samuel, Kevin Coakley et al.

To enhance the reproducibility and reliability of deep learning models, we address a critical gap in current training methodologies: the lack of mechanisms that ensure consistent and robust performance across runs. Our empirical analysis reveals that even under controlled initialization and training conditions, the accuracy of the model can exhibit significant variability. To address this issue, we propose a Custom Loss Function (CLF) that reduces the sensitivity of training outcomes to stochastic factors such as weight initialization and data shuffling. By fine-tuning its parameters, CLF explicitly balances predictive accuracy with training stability, leading to more consistent and reliable model performance. Extensive experiments across diverse architectures for both image classification and time series forecasting demonstrate that our approach significantly improves training robustness without sacrificing predictive performance. These results establish CLF as an effective and efficient strategy for developing more stable, reliable and trustworthy neural networks.

Vamsi Krishna Kommineni, Birgitta König-Ries, Sheeba Samuel

Deep Learning (DL) techniques are increasingly applied in scientific studies across various domains to address complex research questions. However, the methodological details of these DL models are often hidden in the unstructured text. As a result, critical information about how these models are designed, trained, and evaluated is challenging to access and comprehend. To address this issue, in this work, we use five different open-source Large Language Models (LLMs): Llama-3 70B, Llama-3.1 70B, Mixtral-8x22B-Instruct-v0.1, Mixtral 8x7B, and Gemma 2 9B in combination with Retrieval-Augmented Generation (RAG) approach to extract and process DL methodological details from scientific publications automatically. We built a voting classifier from the outputs of five LLMs to accurately report DL methodological information. We tested our approach using biodiversity publications, building upon our previous research. To validate our pipeline, we employed two datasets of DL-related biodiversity publications: a curated set of 100 publications from our prior work and a set of 364 publications from the Ecological Informatics journal. Our results demonstrate that the multi-LLM, RAG-assisted pipeline enhances the retrieval of DL methodological information, achieving an accuracy of 69.5% (417 out of 600 comparisons) based solely on textual content from publications. This performance was assessed against human annotators who had access to code, figures, tables, and other supplementary information. Although demonstrated in biodiversity, our methodology is not limited to this field; it can be applied across other scientific domains where detailed methodological reporting is essential for advancing knowledge and ensuring reproducibility. This study presents a scalable and reliable approach for automating information extraction, facilitating better reproducibility and knowledge transfer across studies.

Sheeba Samuel, Daniel Mietchen, Jungsan Kim et al.

Digital libraries curate millions of research software artefacts yet lack scalable infrastructure for assessing whether those artefacts remain executable. Existing automated assessment tools treat static repository completeness -- what a repository contains -- as a proxy for execution success -- whether it runs. We term this the readiness-outcome conflation and present ReproScore, a two-tier framework that explicitly separates reproducibility readiness (RRS) from reproducibility outcome (ROS), combining them into a coverage-adaptive Composite Score (RCS). RRS comprises 26 sub-metrics across five categories; ROS provides execution-based probes when sandbox infrastructure is available; a community rubric externalises weighting priorities as versioned YAML profiles. Evaluated on 423 GitHub repositories from a large-scale ground-truth corpus spanning five failure modes, two complementary findings emerge: the environment category strongly discriminates failure mode, confirming static signals capture meaningful structural differences; yet RRS exhibits near-zero binary success correlation, empirically quantifying the readiness-outcome gap at repository scale. Together, these findings validate the architectural separation as both necessary and non-trivial, positioning ReproScore as scalable infrastructure for reproducibility-aware curation in digital library workflows.

Jihen Amara, Birgitta König-Ries, Sheeba Samuel

While deep learning has significantly advanced automatic plant disease detection through image-based classification, improving model explainability remains crucial for reliable disease detection. In this study, we apply the Automated Concept-based Explanation (ACE) method to plant disease classification using the widely adopted InceptionV3 model and the PlantVillage dataset. ACE automatically identifies the visual concepts found in the image data and provides insights about the critical features influencing the model predictions. This approach reveals both effective disease-related patterns and incidental biases, such as those from background or lighting that can compromise model robustness. Through systematic experiments, ACE helped us to identify relevant features and pinpoint areas for targeted model improvement. Our findings demonstrate the potential of ACE to improve the explainability of plant disease classification based on deep learning, which is essential for producing transparent tools for plant disease management in agriculture.

Sheeba Samuel, Daniel Mietchen, Hemanta Lo et al.

Computational reproducibility is fundamental to trustworthy science, yet remains difficult to achieve in practice across various research workflows, including Jupyter notebooks published alongside scholarly articles. Environment drift, undocumented dependencies and implicit execution assumptions frequently prevent independent re-execution of published research. Despite existing reproducibility guidelines, scalable and systematic infrastructure for automated assessment remains limited. We present an automated, web-oriented reproducibility engineering pipeline that reconstructs and evaluates repository-level execution environments for scholarly notebooks. The system performs dependency inference, automated container generation, and isolated execution to approximate the notebook's original computational context. We evaluate the approach on 443 notebooks from 116 GitHub repositories referenced by publications in PubMed Central. Execution outcomes are classified into four categories: resolved environment failures, persistent logic or data errors, reproducibility drift, and container-induced regressions. Our results show that containerization resolves 66.7% of prior dependency-related failures and substantially improves execution robustness. However, a significant reproducibility gap remains: 53.7% of notebooks exhibit low output fidelity, largely due to persistent runtime failures and stochastic non-determinism. These findings indicate that standardized containerization is essential for computational stability but insufficient for full bit-wise reproducibility. The framework offers a scalable solution for researchers, editors, and archivists seeking systematic, automated assessment of computational artifacts.

Sheeba Samuel, Frank Löffler, Birgitta König-Ries

Machine learning (ML) is an increasingly important scientific tool supporting decision making and knowledge generation in numerous fields. With this, it also becomes more and more important that the results of ML experiments are reproducible. Unfortunately, that often is not the case. Rather, ML, similar to many other disciplines, faces a reproducibility crisis. In this paper, we describe our goals and initial steps in supporting the end-to-end reproducibility of ML pipelines. We investigate which factors beyond the availability of source code and datasets influence reproducibility of ML experiments. We propose ways to apply FAIR data practices to ML workflows. We present our preliminary results on the role of our tool, ProvBook, in capturing and comparing provenance of ML experiments and their reproducibility using Jupyter Notebooks.