Joint intent detection and slot filling is a key research topic in natural language understanding (NLU). Existing joint intent and slot filling systems analyze and compute features collectively for all slot types, and importantly, have no way to explain the slot filling model decisions. In this work, we propose a novel approach that: (i) learns to generate additional slot type specific features in order to improve accuracy and (ii) provides explanations for slot filling decisions for the first time in a joint NLU model. We perform an additional constrained supervision using a set of binary classifiers for the slot type specific feature learning, thus ensuring appropriate attention weights are learned in the process to explain slot filling decisions for utterances. Our model is inherently explainable and does not need any post-hoc processing. We evaluate our approach on two widely used datasets and show accuracy improvements. Moreover, a detailed analysis is also provided for the exclusive slot explainability.

Machine learning (ML) formalizes the problem of getting computers to learn from experience as optimization of performance according to some metric(s) on a set of data examples. This is in contrast to requiring behaviour specified in advance (e.g. by hard-coded rules). Formalization of this problem has enabled great progress in many applications with large real-world impact, including translation, speech recognition, self-driving cars, and drug discovery. But practical instantiations of this formalism make many assumptions - for example, that data are i.i.d.: independent and identically distributed - whose soundness is seldom investigated. And in making great progress in such a short time, the field has developed many norms and ad-hoc standards, focused on a relatively small range of problem settings. As applications of ML, particularly in artificial intelligence (AI) systems, become more pervasive in the real world, we need to critically examine these assumptions, norms, and problem settings, as well as the methods that have become de-facto standards. There is much we still do not understand about how and why deep networks trained with stochastic gradient descent are able to generalize as well as they do, why they fail when they do, and how they will perform on out-of-distribution data. In this thesis I cover some of my work towards better understanding deep net generalization, identify several ways assumptions and problem settings fail to generalize to the real world, and propose ways to address those failures in practice.

This work presents a novel dense RGB-D SLAM approach for dynamic planar environments that enables simultaneous multi-object tracking, camera localisation and background reconstruction. Previous dynamic SLAM methods either rely on semantic segmentation to directly detect dynamic objects; or assume that dynamic objects occupy a smaller proportion of the camera view than the static background and can, therefore, be removed as outliers. Our approach, however, enables dense SLAM when the camera view is largely occluded by multiple dynamic objects with the aid of camera motion prior. The dynamic planar objects are separated by their different rigid motions and tracked independently. The remaining dynamic non-planar areas are removed as outliers and not mapped into the background. The evaluation demonstrates that our approach outperforms the state-of-the-art methods in terms of localisation, mapping, dynamic segmentation and object tracking. We also demonstrate its robustness to large drift in the camera motion prior.

As computing technology rapidly advances, there has been much discussion of the potential threats posed by artificial intelligence. But the author of a book that explores the nature of machine versus human intelligence has said some of these debates over the future of AI have been overhyped and may be distracting from more pressing issues. Junaid Mubeen, a research mathematician turned educator and author of Mathematical Intelligence: A Story of Human Superiority Over Machines, which will be published November 1, told Newsweek one of the reasons he wrote the book was that AI has generated significant amounts of publicity recently. "Some of it may be justified because there are exciting developments coming through, but much of it, I think, is overhyped," Mubeen said. "And I think there's a real risk that we're going to rush to judgment, exaggerate the capabilities of AI in the process and undermine our own human intelligence. "It was Arthur C. Clarke who said, 'Any sufficiently advanced technology is indistinguishable from magic,' and we're seeing that now," he said. Mubeen pointed to the example of the Google engineer who made headlines earlier this year after saying that a chatbot the company developed, called LaMDA (Language Model for Dialogue Applications), had acquired sentience. At the core of this kind of machine learning-based artificial intelligence, which is being used in a huge variety of applications--everything from medicine and agriculture to astronomy and robotics--is pattern recognition. Machine learning is a subfield of artificial intelligence that gives computers the ability to learn without being explicitly programmed. "That is just one aspect of intelligence," Mubeen said. "They have the appearance of intelligence--when you're engaging with a chatbot, it can feel like you're conversing with a human.

Adrienne Williams and Milagros Miceli are researchers at the Distributed AI Research (DAIR) Institute. Timnit Gebru is the institute's founder and executive director. She was previously co-lead of the Ethical AI research team at Google. The public's understanding of artificial intelligence (AI) is largely shaped by pop culture -- by blockbuster movies like "The Terminator" and their doomsday scenarios of machines going rogue and destroying humanity. This kind of AI narrative is also what grabs the attention of news outlets: a Google engineer claiming that its chatbot was sentient was among the most discussed AI-related news in recent months, even reaching Stephen Colbert's millions of viewers.

You might suppose Hollywood is good at predicting the future. Indeed, Robert Wallace, head of the CIA's Office of Technical Service and the US equivalent of MI6's fictional Q, has recounted how Russian spies would watch the latest Bond movie to see what technologies might be coming their way. Hollywood's continuing obsession with killer robots might therefore be of significant concern. The newest such movie is Apple TV's forthcoming sex robot courtroom drama Dolly. I never thought I'd write the phrase "sex robot courtroom drama", but there you go.

Every now and then you play a video game that you just cannot stop thinking about. Candy Crush might leave colourful imprints on the back of your eyelids. And then, very occasionally, a game comes along that is so entirely unlike anything you've ever played that it becomes an obsession. Immortality, the latest from lauded game-maker Sam Barlow and his studio Half Mermaid, is one of those. It is something that has never existed before: a video game that is also three feature-length films, wrapped around a mystery so compelling that I couldn't concentrate on anything else for days. It is so delicate and complex that it's difficult to figure out how it even works.

Question generation has recently gained a lot of research interest, especially with the advent of large language models. In and of itself, question generation can be considered 'AI-hard', as there is a lack of unanimously agreed sense of what makes a question 'good' or 'bad'. In this paper, we tackle two fundamental problems in parallel: on one hand, we try to solve the scaling problem, where question-generation and answering applications have to be applied to a massive amount of text without ground truth labeling. The usual approach to solve this problem is to either downsample or summarize. However, there are critical risks of misinformation with these approaches. On the other hand, and related to the misinformation problem, we try to solve the 'safety' problem, as many public institutions rely on a much higher level of accuracy for the content they provide. We introduce an adversarial approach to tackle the question generation safety problem with scale. Specifically, we designed a question-answering system that specifically prunes out unanswerable questions that may be generated, and further increases the quality of the answers that are generated. We build a production-ready, easily-plugged pipeline that can be used on any given body of text, that is scalable and immune from generating any hate speech, profanity, or misinformation. Based on the results, we are able to generate more than six times the number of quality questions generated by the abstractive approach, with a perceived quality being 44% higher, according to a survey of 168 participants.

In this paper, we examine several methods of acquiring Czech data for automated fact-checking, which is a task commonly modeled as a classification of textual claim veracity w.r.t. a corpus of trusted ground truths. We attempt to collect sets of data in form of a factual claim, evidence within the ground truth corpus, and its veracity label (supported, refuted or not enough info). As a first attempt, we generate a Czech version of the large-scale FEVER dataset built on top of Wikipedia corpus. We take a hybrid approach of machine translation and document alignment; the approach and the tools we provide can be easily applied to other languages. We discuss its weaknesses and inaccuracies, propose a future approach for their cleaning and publish the 127k resulting translations, as well as a version of such dataset reliably applicable for the Natural Language Inference task - the CsFEVER-NLI. Furthermore, we collect a novel dataset of 3,097 claims, which is annotated using the corpus of 2.2M articles of Czech News Agency. We present its extended annotation methodology based on the FEVER approach, and, as the underlying corpus is kept a trade secret, we also publish a standalone version of the dataset for the task of Natural Language Inference we call CTKFactsNLI. We analyze both acquired datasets for spurious cues - annotation patterns leading to model overfitting. CTKFacts is further examined for inter-annotator agreement, thoroughly cleaned, and a typology of common annotator errors is extracted. Finally, we provide baseline models for all stages of the fact-checking pipeline and publish the NLI datasets, as well as our annotation platform and other experimental data.

Federated learning allows distributed users to collaboratively train a model while keeping each user's data private. Recently, a growing body of work has demonstrated that an eavesdropping attacker can effectively recover image data from gradients transmitted during federated learning. However, little progress has been made in recovering text data. In this paper, we present a novel attack method FILM for federated learning of language models (LMs). For the first time, we show the feasibility of recovering text from large batch sizes of up to 128 sentences. Unlike image-recovery methods that are optimized to match gradients, we take a distinct approach that first identifies a set of words from gradients and then directly reconstructs sentences based on beam search and a prior-based reordering strategy. We conduct the FILM attack on several large-scale datasets and show that it can successfully reconstruct single sentences with high fidelity for large batch sizes and even multiple sentences if applied iteratively. We evaluate three defense methods: gradient pruning, DPSGD, and a simple approach to freeze word embeddings that we propose. We show that both gradient pruning and DPSGD lead to a significant drop in utility. However, if we fine-tune a public pre-trained LM on private text without updating word embeddings, it can effectively defend the attack with minimal data utility loss. Together, we hope that our results can encourage the community to rethink the privacy concerns of LM training and its standard practices in the future.