Authors pictured in order of their interview publication date (left to right, top to bottom). Each year, a small group of PhD students are chosen to participate in the AAAI/SIGAI Doctoral Consortium . This initiative provides an opportunity for the students to discuss and explore their research interests and career objectives in an interdisciplinary workshop together with a panel of established researchers. During 2025, we met with some of the students to find out more about their research and the doctoral consortium experience. Kunpeng Xu completed his PhD at the Université de Sherbrooke and is now a postdoctoral fellow at McGill University.

In this interview series, we're meeting some of the AAAI/SIGAI Doctoral Consortium participants to find out more about their research. Janice Anta Zebaze is using AI to address energy supply challenges and she told us more about the research she's carried our so far, her plans for further investigations, and what inspired her to pursue a PhD in the field. Tell us a bit about your PhD - where are you studying, and what is the topic of your research? I am currently pursuing my PhD in Physics at the University of Yaounde I in Cameroon, with a focus on renewable energy systems, tribology, and artificial intelligence. The aim of my research is to address energy supply challenges in developing countries by leveraging AI to evaluate resource availability and optimize energy systems.

Scholarly work and communication encompass the entire system in which research and creative works are created, evaluated for quality, disseminated to the academic community and beyond, used, and preserved for future use. It includes formal publications, such as journal articles and books, as well as informal sharing through preprints, conference presentations, data sharing, and broader engagement with scholarly works and research outputs. Scholarly knowledge serves as the primary engine of progress, shaping our world and guiding our collective future. It forms the backbone of technological advancement, public health systems, and sustainable environmental practices. Obtained through rigorous methods of observation, experimentation, and validation, it is a reliable resource that helps societies solve complex problems and improve the quality of life by achieving sustainable development goals (SDGs) [6]. To be truly useful, scholarly knowledge must first be systematically extracted and organized. However, the scholarly community of today faces the problem of an overload of scientific papers in their respective domains. There is an increasing number of papers published every year (currently, 3 million), in addition to more than 200 million papers that have already been published . This gives rise to the research question: "How can we provide a reliable and living scholarly knowledge base that empowers researchers to query, synthesize, and analyze the vast body of scholarly knowledge?"

Deep learning has emerged as a transformative technology in agricultural science, particularly for the identification of plant diseases. This approach leverages advanced algorithms, primarily Convolutional Neural Networks (CNNs), to analyze images of plants and accurately diagnose diseases that threaten crop health and yield (Mohanty et al., 2016; Guo et al., 2020; Saleem et al., 2020; Ahmed & Y adav, 2023; Jung et al., 2023; Shoaib et al., 2023; Pacal et al., 2024). Plant diseases pose a significant threat to global food security, leading to substantial yield losses and economic impacts on agriculture. T raditional methods of disease identification often rely on visual assessments by trained professionals, which can be time-consuming, subjective, and prone to errors (Jafar et al., 2024). As a result, there is a pressing need for automated systems that can provide rapid and accurate disease detection to support farmers and agricultural experts in managing crop health effectively . Deep learning models, especially CNNs, have been shown to outperform traditional methods in terms of accuracy and efficiency . These models can learn hierarchical representations from raw image data, enabling them to identify complex patterns associated with various plant diseases. Recent studies have demonstrated that CNNs can achieve accuracy rates as high as 99.35% when classifying images of diseased and healthy plants. The architecture of CNNs typically includes layers for feature extraction and classification, allowing them to process visual information effectively .

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 3,000 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai.

Interpretability for Table Question Answering (Table QA) is critical, particularly in high-stakes industries like finance or healthcare. Although recent approaches using Large Language Models (LLMs) have significantly improved Table QA performance, their explanations for how the answers are generated are ambiguous. To fill this gap, we introduce Plan-of-SQLs ( or POS), an interpretable, effective, and efficient approach to Table QA that answers an input query solely with SQL executions. Through qualitative and quantitative evaluations with human and LLM judges, we show that POS is most preferred among explanation methods, helps human users understand model decision boundaries, and facilitates model success and error identification. Furthermore, when evaluated in standard benchmarks (TabFact, WikiTQ, and FetaQA), POS achieves competitive or superior accuracy compared to existing methods, while maintaining greater efficiency by requiring significantly fewer LLM calls and database queries.

The Scholarly Hybrid Question Answering over Linked Data (QALD) Challenge at the International Semantic Web Conference (ISWC) 2024 focuses on Question Answering (QA) over diverse scholarly sources: DBLP, SemOpenAlex, and Wikipedia-based texts. This paper describes a methodology that combines SPARQL queries, divide and conquer algorithms, and a pre-trained extractive question answering model. It starts with SPARQL queries to gather data, then applies divide and conquer to manage various question types and sources, and uses the model to handle personal author questions. The approach, evaluated with Exact Match and F-score metrics, shows promise for improving QA accuracy and efficiency in scholarly contexts. Keywords: Scholarly Question Answering, Large Language Models, Divide and conquer.

We investigate the internal representations of vision-and-language models (VLMs) and how they encode task representations. We consider tasks specified through examples or instructions, using either text or image inputs. Surprisingly, we find that conceptually similar tasks are mapped to similar task vector representations, regardless of how they are specified. Our findings suggest that to output answers, tokens in VLMs undergo three distinct phases: input, task, and answer, a process which is consistent across different modalities and specifications. The task vectors we identify in VLMs are general enough to be derived in one modality (e.g., text) and transferred to another (e.g., image). Additionally, we find that ensembling exemplar and instruction based task vectors produce better task representations. Taken together, these insights shed light on the underlying mechanisms of VLMs, particularly their ability to represent tasks in a shared manner across different modalities and task specifications. Project page: https://task-vectors-are-cross-modal.github.io.

Data mining, particularly the analysis of multivariate time series data, plays a crucial role in extracting insights from complex systems and supporting informed decision-making across diverse domains. However, assessing the similarity of multivariate time series data presents several challenges, including dealing with large datasets, addressing temporal misalignments, and the need for efficient and comprehensive analytical frameworks. To address all these challenges, we propose a novel integrated computational approach known as Multivariate Time series Alignment and Similarity Assessment (MTASA). MTASA is built upon a hybrid methodology designed to optimize time series alignment, complemented by a multiprocessing engine that enhances the utilization of computational resources. This integrated approach comprises four key components, each addressing essential aspects of time series similarity assessment, thereby offering a comprehensive framework for analysis. MTASA is implemented as an open-source Python library with a user-friendly interface, making it accessible to researchers and practitioners. To evaluate the effectiveness of MTASA, we conducted an empirical study focused on assessing agroecosystem similarity using real-world environmental data. The results from this study highlight MTASA's superiority, achieving approximately 1.5 times greater accuracy and twice the speed compared to existing state-of-the-art integrated frameworks for multivariate time series similarity assessment. It is hoped that MTASA will significantly enhance the efficiency and accessibility of multivariate time series analysis, benefitting researchers and practitioners across various domains. Its capabilities in handling large datasets, addressing temporal misalignments, and delivering accurate results make MTASA a valuable tool for deriving insights and aiding decision-making processes in complex systems.

Machine learning techniques, such as deep learning and ensemble methods, are widely used in various domains due to their ability to handle complex real-world tasks. However, their black-box nature has raised multiple concerns about the fairness, trustworthiness, and transparency of computer-assisted decision-making. This has led to the emergence of local post-hoc explainability methods, which offer explanations for individual decisions made by black-box algorithms. Among these methods, Kernel SHAP is widely used due to its model-agnostic nature and its well-founded theoretical framework. Despite these strengths, Kernel SHAP suffers from high instability: different executions of the method with the same inputs can lead to significantly different explanations, which diminishes the utility of post-hoc explainability. The contribution of this paper is two-fold. On the one hand, we show that Kernel SHAP's instability is caused by its stochastic neighbor selection procedure, which we adapt to achieve full stability without compromising explanation fidelity. On the other hand, we show that by restricting the neighbors generation to perturbations of size 1 -- which we call the coalitions of Layer 1 -- we obtain a novel feature-attribution method that is fully stable, efficient to compute, and still meaningful.