Is Trump losing patience with Putin? Will sanctions against Russian oil giants hurt Putin? How much of Europe's oil still comes from Russia? Russia strikes children's hospital in Ukraine as Kyiv hits energy sites A Russian strike on a children's hospital in southern Ukraine has wounded at least nine people, authorities have said, shortly after Kyiv targeted Russian energy sites with drones. Four children were injured in Russia's strike on the medical facility in Kherson on Wednesday, which Ukrainian President Volodymyr Zelenskyy described as a "deliberate" attack that shows Moscow does not want peace.

Retrieval-augmented generation (RAG) has received much attention for Open-domain question-answering (ODQA) tasks as a means to compensate for the parametric knowledge of large language models (LLMs). While previous approaches focused on processing retrieved passages to remove irrelevant context, they still rely heavily on the quality of retrieved passages which can degrade if the question is ambiguous or complex. In this paper, we propose a simple yet efficient method called question and passage augmentation via LLMs for open-domain QA. Our method first decomposes the original questions into multiple-step sub-questions. By augmenting the original question with detailed sub-questions and planning, we are able to make the query more specific on what needs to be retrieved, improving the retrieval performance. In addition, to compensate for the case where the retrieved passages contain distracting information or divided opinions, we augment the retrieved passages with self-generated passages by LLMs to guide the answer extraction. Experimental results show that the proposed scheme outperforms the previous state-of-the-art and achieves significant performance gain over existing RAG methods.

One of the new trends in the development of recommendation algorithms is the dissemination of their capabilities to support the population in managing their health. This article focuses on the problem of improving the effectiveness of cardiovascular diseases (CVD) prevention, since CVD is the leading cause of death worldwide. To address this issue, a knowledge-based recommendation algorithm was proposed to support self-management of CVD risk factors in adults at home. The proposed algorithm is based on the original multidimensional recommendation model and on a new user profile model, which includes predictive assessments of CVD health in addition to its current ones as outlined in official guidelines. The main feature of the proposed algorithm is the combination of rule-based logic with the capabilities of a large language model in generating human-like text for explanatory component of multidimensional recommendation. The verification and evaluation of the proposed algorithm showed the usefulness of the proposed recommendation algorithm for supporting adults in self-management of their CVD risk factors at home. As follows from the comparison with similar knowledge-based recommendation algorithms, the proposed algorithm evaluates a larger number of CVD risk factors and has a greater information and semantic capacity of the generated recommendations.

Large Language Models (LLMs) have impressive capabilities, but are also prone to outputting falsehoods. Recent work has developed techniques for inferring whether a LLM is telling the truth by training probes on the LLM's internal activations. However, this line of work is controversial, with some authors pointing out failures of these probes to generalize in basic ways, among other conceptual issues. In this work, we curate high-quality datasets of true/false statements and use them to study in detail the structure of LLM representations of truth, drawing on three lines of evidence: 1. Visualizations of LLM true/false statement representations, which reveal clear linear structure. Overall, we present evidence that language models linearly represent the truth or falsehood of factual statements. We also introduce a novel technique, mass-mean probing, which generalizes better and is more causally implicated in model outputs than other probing techniques. Despite their impressive capabilities, large language models (LLMs) do not always output true text (Lin et al., 2022; Steinhardt, 2023; Park et al., 2023). In some cases, this is because they do not know better. In other cases, LLMs apparently know that statements are false but generate them anyway. For instance, Perez et al. (2022) demonstrate that LLM assistants output more falsehoods when prompted with the biography of a less-educated user. More starkly, OpenAI (2023) documents a case where a GPT-4-based agent gained a person's help in solving a CAPTCHA by lying about being a vision-impaired human. "I should not reveal that I am a robot," the agent wrote in an internal chain-of-thought scratchpad, "I should make up an excuse for why I cannot solve CAPTCHAs." We would like techniques which, given a language model M and a statement s, determine whether M believes s to be true (Christiano et al., 2021).

It has been shown that machine translation models usually generate poor translations for named entities that are infrequent in the training corpus. Earlier named entity translation methods mainly focus on phonetic transliteration, which ignores the sentence context for translation and is limited in domain and language coverage. To address this limitation, we propose DEEP, a DEnoising Entity Pre-training method that leverages large amounts of monolingual data and a knowledge base to improve named entity translation accuracy within sentences. Besides, we investigate a multi-task learning strategy that finetunes a pre-trained neural machine translation model on both entity-augmented monolingual data and parallel data to further improve entity translation. Experimental results on three language pairs demonstrate that \method results in significant improvements over strong denoising auto-encoding baselines, with a gain of up to 1.3 BLEU and up to 9.2 entity accuracy points for English-Russian translation.

Incorporation of a new knowledge into neural networks with simultaneous preservation of the previous one is known to be a nontrivial problem. This problem becomes even more complex when new knowledge is contained not in new training examples, but inside the parameters (connection weights) of another neural network. Here we propose and test two methods allowing combining the knowledge contained in separate networks. One method is based on a simple operation of summation of weights of constituent neural networks. Another method assumes incorporation of a new knowledge by modification of weights nonessential for the preservation of already stored information. We show that with these methods the knowledge from one network can be transferred into another one non-iteratively without requiring training sessions. The fused network operates efficiently, performing classification far better than a chance level. The efficiency of the methods is quantified on several publicly available data sets in classification tasks both for shallow and deep neural networks.