We develop a general framework for abstracting the behavior of an agent that operates in a nondeterministic domain, i.e., where the agent does not control the outcome of the nondeterministic actions, based on the nondeterministic situation calculus and the ConGolog programming language. We assume that we have both an abstract and a concrete nondeterministic basic action theory, and a refinement mapping which specifies how abstract actions, decomposed into agent actions and environment reactions, are implemented by concrete ConGolog programs. This new setting supports strategic reasoning and strategy synthesis, by allowing us to quantify separately on agent actions and environment reactions. We show that if the agent has a (strong FOND) plan/strategy to achieve a goal/complete a task at the abstract level, and it can always execute the nondeterministic abstract actions to completion at the concrete level, then there exists a refinement of it that is a (strong FOND) plan/strategy to achieve the refinement of the goal/task at the concrete level.

Ensembling BERT models often significantly improves accuracy, but at the cost of significantly more computation and memory footprint. In this work, we propose Multi-CLS BERT, a novel ensembling method for CLS-based prediction tasks that is almost as efficient as a single BERT model. Multi-CLS BERT uses multiple CLS tokens with a parameterization and objective that encourages their diversity. Thus instead of fine-tuning each BERT model in an ensemble (and running them all at test time), we need only fine-tune our single Multi-CLS BERT model (and run the one model at test time, ensembling just the multiple final CLS embeddings). To test its effectiveness, we build Multi-CLS BERT on top of a state-of-the-art pretraining method for BERT (Aroca-Ouellette and Rudzicz, 2020). In experiments on GLUE and SuperGLUE we show that our Multi-CLS BERT reliably improves both overall accuracy and confidence estimation. When only 100 training samples are available in GLUE, the Multi-CLS BERT_Base model can even outperform the corresponding BERT_Large model. We analyze the behavior of our Multi-CLS BERT, showing that it has many of the same characteristics and behavior as a typical BERT 5-way ensemble, but with nearly 4-times less computation and memory.

The VNHSGE (VietNamese High School Graduation Examination) dataset, developed exclusively for evaluating large language models (LLMs), is introduced in this article. The dataset, which covers nine subjects, was generated from the Vietnamese National High School Graduation Examination and comparable tests. 300 literary essays have been included, and there are over 19,000 multiple-choice questions on a range of topics. The dataset assesses LLMs in multitasking situations such as question answering, text generation, reading comprehension, visual question answering, and more by including both textual data and accompanying images. Using ChatGPT and BingChat, we evaluated LLMs on the VNHSGE dataset and contrasted their performance with that of Vietnamese students to see how well they performed. The results show that ChatGPT and BingChat both perform at a human level in a number of areas, including literature, English, history, geography, and civics education. They still have space to grow, though, especially in the areas of mathematics, physics, chemistry, and biology. The VNHSGE dataset seeks to provide an adequate benchmark for assessing the abilities of LLMs with its wide-ranging coverage and variety of activities. We intend to promote future developments in the creation of LLMs by making this dataset available to the scientific community, especially in resolving LLMs' limits in disciplines involving mathematics and the natural sciences.

After ChatGPT burst on the scene last November, some government officials raced to prohibit its use. New York City, Los Angeles Unified, Seattle, and Baltimore School Districts either banned or blocked access to generative AI tools, fearing that ChatGPT, Bard, and other content generation sites could tempt students to cheat on assignments, induce rampant plagiarism, and impede critical thinking. This week, US Congress heard testimony from Sam Altman, CEO of OpenAI, and AI researcher Gary Marcus as it weighed whether and how to regulate the technology. In a rapid about-face, however, a few governments are now embracing a less fearful and more hands-on approach to AI. New York City Schools chancellor David Banks announced yesterday that NYC is reversing its ban because "the knee jerk fear and risk overlooked the potential of generative AI to support students and teachers, as well as the reality that our students are participating in and will work in a world where understanding generative AI is crucial."

Modern approaches for keyword spotting rely on training deep neural networks on large static datasets with i.i.d. distributions. However, the resulting models tend to underperform when presented with changing data regimes in real-life applications. This work investigates a simple but effective online continual learning method that updates a keyword spotter on-device via SGD as new data becomes available. Contrary to previous research, this work focuses on learning the same KWS task, which covers most commercial applications. During experiments with dynamic audio streams in different scenarios, that method improves the performance of a pre-trained small-footprint model by 34%. Moreover, experiments demonstrate that, compared to a naive online learning implementation, conditional model updates based on its performance in a small hold-out set drawn from the training distribution mitigate catastrophic forgetting.

A basic task for most Legal Artificial Intelligence (Legal AI) applications is Named Entity Recognition (NER). However, texts produced in the context of legal practice make references to entities that are not trivially recognized by the currently available NERs. There is a lack of categorization of legislation, jurisprudence, evidence, penalties, the roles of people in a legal process (judge, lawyer, victim, defendant, witness), types of locations (crime location, defendant's address), etc. In this sense, there is still a need for a robust golden collection, annotated with fine-grained entities of the legal domain, and which covers various documents of a legal process, such as petitions, inquiries, complaints, decisions and sentences. In this article, we describe the development of the Golden Collection of the Brazilian Judiciary (CDJUR-BR) contemplating a set of fine-grained named entities that have been annotated by experts in legal documents. The creation of CDJUR-BR followed its own methodology that aimed to attribute a character of comprehensiveness and robustness. Together with the CDJUR-BR repository we provided a NER based on the BERT model and trained with the CDJUR-BR, whose results indicated the prevalence of the CDJUR-BR.

Machine learning is employed in healthcare to draw approximate conclusions regarding human diseases and mental health problems. Compared to older traditional methods, it can help to analyze data more efficiently and produce better and more dependable results. Millions of people are affected by schizophrenia, which is a chronic mental disorder that can significantly impact their lives. Many machine learning algorithms have been developed to predict and prevent this disease, and they can potentially be implemented in the diagnosis of individuals who have it. This survey aims to review papers that have focused on the use of deep learning to detect and predict schizophrenia using EEG signals, functional magnetic resonance imaging (fMRI), and diffusion magnetic resonance imaging (dMRI). With our chosen search strategy, we assessed ten publications from 2019 to 2022. All studies achieved successful predictions of more than 80%. This review provides summaries of the studies and compares their notable aspects. In the field of artificial intelligence (AI) and machine learning (ML) for schizophrenia, significant advances have been made due to the availability of ML tools, and we are optimistic that this field will continue to grow.

Inferring a state sequence from a sequence of measurements is a fundamental problem in bioinformatics and natural language processing. The Viterbi and the Beam Search (BS) algorithms are popular inference methods, but they have limitations when applied to Hierarchical Hidden Markov Models (HHMMs), where the interest lies in the outer state sequence. The Viterbi algorithm can not infer outer states without inner states, while the BS algorithm requires marginalization over prohibitively large state spaces. We propose two new algorithms to overcome these limitations: the greedy marginalized BS algorithm and the local focus BS algorithm. We show that they approximate the most likely outer state sequence with higher performance than the Viterbi algorithm, and we evaluate the performance of these algorithms on an explicit duration HMM with simulation and nanopore base calling data.

We study the computational complexity of counterfactual reasoning in relation to the complexity of associational and interventional reasoning on structural causal models (SCMs). We show that counterfactual reasoning is no harder than associational or interventional reasoning on fully specified SCMs in the context of two computational frameworks. The first framework is based on the notion of treewidth and includes the classical variable elimination and jointree algorithms. The second framework is based on the more recent and refined notion of causal treewidth which is directed towards models with functional dependencies such as SCMs. Our results are constructive and based on bounding the (causal) treewidth of twin networks -- used in standard counterfactual reasoning that contemplates two worlds, real and imaginary -- to the (causal) treewidth of the underlying SCM structure. In particular, we show that the latter (causal) treewidth is no more than twice the former plus one. Hence, if associational or interventional reasoning is tractable on a fully specified SCM then counterfactual reasoning is tractable too. We extend our results to general counterfactual reasoning that requires contemplating more than two worlds and discuss applications of our results to counterfactual reasoning with a partially specified SCM that is coupled with data. We finally present empirical results that measure the gap between the complexities of counterfactual reasoning and associational/interventional reasoning on random SCMs.

The war in Ukraine seems to have positively changed the attitude toward the critical societal topic of migration in Europe -- at least towards refugees from Ukraine. We investigate whether this impression is substantiated by how the topic is reflected in online news and social media, thus linking the representation of the issue on the Web to its perception in society. For this purpose, we combine and adapt leading-edge automatic text processing for a novel multilingual stance detection approach. Starting from 5.5M Twitter posts published by 565 European news outlets in one year, beginning September 2021, plus replies, we perform a multilingual analysis of migration-related media coverage and associated social media interaction for Europe and selected European countries. The results of our analysis show that there is actually a reframing of the discussion illustrated by the terminology change, e.g., from "migrant" to "refugee", often even accentuated with phrases such as "real refugees". However, concerning a stance shift in public perception, the picture is more diverse than expected. All analyzed cases show a noticeable temporal stance shift around the start of the war in Ukraine. Still, there are apparent national differences in the size and stability of this shift.