A plethora of recent research has proposed several automated methods based on machine learning (ML) and deep learning (DL) to detect cybersickness in Virtual reality (VR). However, these detection methods are perceived as computationally intensive and black-box methods. Thus, those techniques are neither trustworthy nor practical for deploying on standalone VR head-mounted displays (HMDs). This work presents an explainable artificial intelligence (XAI)-based framework VR-LENS for developing cybersickness detection ML models, explaining them, reducing their size, and deploying them in a Qualcomm Snapdragon 750G processor-based Samsung A52 device. Specifically, we first develop a novel super learning-based ensemble ML model for cybersickness detection. Next, we employ a post-hoc explanation method, such as SHapley Additive exPlanations (SHAP), Morris Sensitivity Analysis (MSA), Local Interpretable Model-Agnostic Explanations (LIME), and Partial Dependence Plot (PDP) to explain the expected results and identify the most dominant features. The super learner cybersickness model is then retrained using the identified dominant features. Our proposed method identified eye tracking, player position, and galvanic skin/heart rate response as the most dominant features for the integrated sensor, gameplay, and bio-physiological datasets. We also show that the proposed XAI-guided feature reduction significantly reduces the model training and inference time by 1.91X and 2.15X while maintaining baseline accuracy. For instance, using the integrated sensor dataset, our reduced super learner model outperforms the state-of-the-art works by classifying cybersickness into 4 classes (none, low, medium, and high) with an accuracy of 96% and regressing (FMS 1-10) with a Root Mean Square Error (RMSE) of 0.03.

Large Language Models (LLMs) handle physical commonsense information inadequately. As a result of being trained in a disembodied setting, LLMs often fail to predict an action's outcome in a given environment. However, predicting the effects of an action before it is executed is crucial in planning, where coherent sequences of actions are often needed to achieve a goal. Therefore, we introduce the multi-modal task of predicting the outcomes of actions solely from realistic sensory inputs (images and text). Next, we extend an LLM to model latent representations of objects to better predict action outcomes in an environment. We show that multi-modal models can capture physical commonsense when augmented with visual information. Finally, we evaluate our model's performance on novel actions and objects and find that combining modalities help models to generalize and learn physical commonsense reasoning better.

Due to limitations in the computational resources available, General Circulation Models (GCMs) are advocated to simulate the climate system over coarse resolution grids. This hampers the applicability of GCM products in the regional-to-local scale, highly demanded by different socio-economic sectors. Statistical downscaling aims to solve this problem by generating high-resolution climate fields. Recently, machine learning techniques (particularly deep learning models) have shown promising results in this task. These models are first trained in a historical period through observational datasets, and then applied to the GCM outputs of plausible far-future scenarios, thus generating high-resolution climate change products. To assess the plausibility of the derived downscaled fields, several validation frameworks are performed, (e.g., skill to reproduce the present climate) which aim to assess the generalization of the models. Here, we present a novel evaluation protocol building on eXplainable Artificial Intelligence (XAI) to examine the suitability of certain deep learning models for climate downscaling.

We propose a novel high-fidelity face swapping method called "Arithmetic Face Swapping" (AFS) that explicitly disentangles the intermediate latent space W+ of a pretrained StyleGAN into the "identity" and "style" subspaces so that a latent code in W+ is the sum of an "identity" code and a "style" code in the corresponding subspaces. Via our disentanglement, face swapping (FS) can be regarded as a simple arithmetic operation in W+, i.e., the summation of a source "identity" code and a target "style" code. This makes AFS more intuitive and elegant than other FS methods. In addition, our method can generalize over the standard face swapping to support other interesting operations, e.g., combining the identity of one source with styles of multiple targets and vice versa. We implement our identity-style disentanglement by learning a neural network that maps a latent code to a "style" code. We provide a condition for this network which theoretically guarantees identity preservation of the source face even after a sequence of face swapping operations. Extensive experiments demonstrate the advantage of our method over state-of-the-art FS methods in producing high-quality swapped faces. Our source code was made public at https://github.com/truongvu2000nd/AFS

We propose a methodology and design two benchmark sets for measuring to what extent language-and-vision language models use the visual signal in the presence or absence of stereotypes. The first benchmark is designed to test for stereotypical colors of common objects, while the second benchmark considers gender stereotypes. The key idea is to compare predictions when the image conforms to the stereotype to predictions when it does not. Our results show that there is significant variation among multimodal models: the recent Transformer-based FLAVA seems to be more sensitive to the choice of image and less affected by stereotypes than older CNN-based models such as VisualBERT and LXMERT. This effect is more discernible in this type of controlled setting than in traditional evaluations where we do not know whether the model relied on the stereotype or the visual signal.

Recently, contrastive learning attracts increasing interests in neural text generation as a new solution to alleviate the exposure bias problem. It introduces a sequence-level training signal which is crucial to generation tasks that always rely on auto-regressive decoding. However, previous methods using contrastive learning in neural text generation usually lead to inferior performance. In this paper, we analyse the underlying reasons and propose a new Contrastive Neural Text generation framework, CoNT. CoNT addresses bottlenecks that prevent contrastive learning from being widely adopted in generation tasks from three aspects -- the construction of contrastive examples, the choice of the contrastive loss, and the strategy in decoding. We validate CoNT on five generation tasks with ten benchmarks, including machine translation, summarization, code comment generation, data-to-text generation and commonsense generation. Experimental results show that CoNT clearly outperforms the conventional training framework on all the ten benchmarks with a convincing margin. Especially, CoNT surpasses previous the most competitive contrastive learning method for text generation, by 1.50 BLEU on machine translation and 1.77 ROUGE-1 on summarization, respectively. It achieves new state-of-the-art on summarization, code comment generation (without external data) and data-to-text generation.

This paper describes the development of an on-board data-driven system that can monitor and localize the fault in a quadrotor unmanned aerial vehicle (UAV) and at the same time, evaluate the degree of damage of the fault under real scenarios. To achieve offline training data generation, a hybrid approach is proposed for the development of a virtual data-generative model using a combination of data-driven models as well as well-established dynamic models that describe the kinematics of the UAV. To effectively represent the drop in performance of a faulty propeller, a variation of the deep neural network, a LSTM network is proposed. With the RPM of the propeller as input and based on the fault condition of the propeller, the proposed propeller model estimates the resultant torque and thrust. Then, flight datasets of the UAV under various fault scenarios are generated via simulation using the developed data-generative model. Lastly, a fault classifier using a CNN model is proposed to identify as well as evaluate the degree of damage to the damaged propeller. The scope of this paper focuses on the identification of faulty propellers and classification of the fault level for quadrotor UAVs using RPM as well as flight data. Doing so allows for early minor fault detection to prevent serious faults from occurring if the fault is left unrepaired. To further validate the workability of this approach outside of simulation, a real-flight test is conducted indoors. The real flight data is collected and a simulation to real sim-real test is conducted. Due to the imperfections in the build of our experimental UAV, a slight calibration approach to our simulation model is further proposed and the experimental results obtained show that our trained model can identify the location of propeller fault as well as the degree/type of damage. Currently, the diagnosis accuracy on the testing set is over 80%.

Given a sentence "Abby told Brittney that she upset Courtney", one would struggle to understand who "she" refers to, and ask for clarification. However, if the word "upset" were replaced with "hugged", "she" unambiguously refers to Abby. We study if modern coreference resolution models are sensitive to such pronominal ambiguity. To this end, we construct AmbiCoref, a diagnostic corpus of minimal sentence pairs with ambiguous and unambiguous referents. Our examples generalize psycholinguistic studies of human perception of ambiguity around particular arrangements of verbs and their arguments. Analysis shows that (1) humans are less sure of referents in ambiguous AmbiCoref examples than unambiguous ones, and (2) most coreference models show little difference in output between ambiguous and unambiguous pairs. We release AmbiCoref as a diagnostic corpus for testing whether models treat ambiguity similarly to humans.

To act in the world, robots rely on a representation of salient task aspects: for example, to carry a cup of coffee, a robot must consider movement efficiency and cup orientation in its behaviour. However, if we want robots to act for and with people, their representations must not be just functional but also reflective of what humans care about, i.e. their representations must be aligned with humans'. In this survey, we pose that current reward and imitation learning approaches suffer from representation misalignment, where the robot's learned representation does not capture the human's representation. We suggest that because humans will be the ultimate evaluator of robot performance in the world, it is critical that we explicitly focus our efforts on aligning learned task representations with humans, in addition to learning the downstream task. We advocate that current representation learning approaches in robotics should be studied from the perspective of how well they accomplish the objective of representation alignment. To do so, we mathematically define the problem, identify its key desiderata, and situate current robot learning methods within this formalism. We conclude the survey by suggesting future directions for exploring open challenges.

Knowledge of syntax includes knowledge of rare, idiosyncratic constructions. LLMs must overcome frequency biases in order to master such constructions. In this study, I prompt GPT-3 to give acceptability judgments on the English-language Article + Adjective + Numeral + Noun construction (e.g., "a lovely five days"). I validate the prompt using the CoLA corpus of acceptability judgments and then zero in on the AANN construction. I compare GPT- 3's judgments to crowdsourced human judgments on a subset of sentences. GPT-3's judgments are broadly similar to human judgments and generally align with proposed constraints in the literature but, in some cases, GPT-3's judgments and human judgments diverge from the literature and from each other.