Retrieval-Augmented Generation (RAG) shows impressive performance by supplementing and substituting parametric knowledge in Large Language Models (LLMs). Retrieved knowledge can be divided into three types: explicit answer evidence, implicit answer clue, and insufficient answer context which can be further categorized into totally irrelevant and partially relevant information. Effectively utilizing partially relevant knowledge remains a key challenge for RAG systems, especially in incomplete knowledge base retrieval. Contrary to the conventional view, we propose a new perspective: LLMs can be awakened via partially relevant knowledge already embedded in LLMs. To comprehensively investigate this phenomenon, the triplets located in the gold reasoning path and their variants are used to construct partially relevant knowledge by removing the path that contains the answer. We provide theoretical analysis of the awakening effect in LLMs and support our hypothesis with experiments on two Knowledge Graphs (KGs) Question Answering (QA) datasets. Furthermore, we present a new task, Unseen Entity KGQA, simulating real-world challenges where entity linking fails due to KG incompleteness. Our awakening-based approach demonstrates greater efficacy in practical applications, outperforms traditional methods that rely on embedding-based similarity which are prone to returning noisy information.

Artificial Intelligence (AI) systems are threaded throughout modern society, informing us in low-risk interactions such as movie recommendations and chatbots to high-risk environments like medical diagnosis, self-driving cars, drones, and military operations. But is remains a significant challenge to develop human trust in these systems, particularly because the systems themselves cannot explain in a way graspable to humans how a recommendation or decision was reached. This lack of trust can become problematic in critical situations involving finances or healthcare where AI decisions can have life-altering consequences. To address this issue, eXplainable Artificial Intelligence (XAI) has become an active research area both for scientists and industry. XAI develops models using explanations that aim to shed light on the underlying mechanisms of AI systems, thus bringing transparency to the process.

[This paper was initially published in PHME conference in 2016, selected for further publication in International Journal of Prognostics and Health Management.] This paper describes an Autoregressive Partially-hidden Markov model (ARPHMM) for fault detection and prognostics of equipments based on sensors' data. It is a particular dynamic Bayesian network that allows to represent the dynamics of a system by means of a Hidden Markov Model (HMM) and an autoregressive (AR) process. The Markov chain assumes that the system is switching back and forth between internal states while the AR process ensures a temporal coherence on sensor measurements. A sound learning procedure of standard ARHMM based on maximum likelihood allows to iteratively estimate all parameters simultaneously. This paper suggests a modification of the learning procedure considering that one may have prior knowledge about the structure which becomes partially hidden. The integration of the prior is based on the Theory of Weighted Distributions which is compatible with the Expectation-Maximization algorithm in the sense that the convergence properties are still satisfied. We show how to apply this model to estimate the remaining useful life based on health indicators. The autoregressive parameters can indeed be used for prediction while the latent structure can be used to get information about the degradation level. The interest of the proposed method for prognostics and health assessment is demonstrated on CMAPSS datasets.

Influence Maximization is an active topic, but it was always assumed full knowledge of the social network graph. However, the graph may actually be unknown beforehand. For example, when selecting a subset of a homeless population to attend interventions concerning health, we deal with a network that is not fully known. Hence, we introduce the novel problem of simultaneously influencing and mapping (i.e., learning) the graph. We study a class of algorithms, where we show that: (i) traditional algorithms may have arbitrarily low performance; (ii) we can effectively influence and map when the independence of objectives hypothesis holds; (iii) when it does not hold, the upper bound for the influence loss converges to 0. We run extensive experiments over four real-life social networks, where we study two alternative models, and obtain significantly better results in both than traditional approaches.

Arriving at the complete probabilistic knowledge of a domain, i.e., learning how all variables interact, is indeed a demanding task. In reality, settings often arise for which an individual merely possesses partial knowledge of the domain, and yet, is expected to give adequate answers to a variety of posed queries. That is, although precise answers to some queries, in principle, cannot be achieved, a range of plausible answers is attainable for each query given the available partial knowledge. In this paper, we propose the Multi-Context Model (MCM), a new graphical model to represent the state of partial knowledge as to a domain. MCM is a middle ground between Probabilistic Logic, Bayesian Logic, and Probabilistic Graphical Models. For this model we discuss: (i) the dynamics of constructing a contradiction-free MCM, i.e., to form partial beliefs regarding a domain in a gradual and probabilistically consistent way, and (ii) how to perform inference, i.e., to evaluate a probability of interest involving some variables of the domain.