Dialog systems, such as voice assistants, are expected to engage with users in complex, evolving conversations. Unfortunately, traditional automatic speech recognition (ASR) systems deployed in such applications are usually trained to recognize each turn independently and lack the ability to adapt to the conversational context or incorporate user feedback. In this work, we introduce a general framework for ASR in dialog systems that can go beyond learning from single-turn utterances and learn over time how to adapt to both explicit supervision and implicit user feedback present in multi-turn conversations. We accomplish that by leveraging advances in student-teacher learning and context-aware dialog processing, and designing contrastive self-supervision approaches with Ohm, a new online hard-negative mining approach. We show that leveraging our new framework compared to traditional training leads to relative WER reductions of close to 10% in real-world dialog systems, and up to 26% on public synthetic data.

One of the most useful applications of intelligent aerial robots sometimes called Unmanned Aerial Vehicles (UAV) in Australia is known to be in bushfire monitoring and prediction operations. A swarm of autonomous drones/UAVs programmed to work in real-time observing the fire parameters using their onboard sensors would be valuable in reducing the life-threatening impact of that fire. However autonomous UAVs face serious challenges in their positioning and navigation in critical bushfire conditions such as remoteness and severe weather conditions where GPS signals could also be unreliable. This paper tackles one of the most important factors in autonomous UAV navigation, namely Initial Positioning sometimes called Localisation. The solution provided by this paper will enable a team of autonomous UAVs to establish a relative position to their base of operation to be able to commence a team search and reconnaissance in a bushfire-affected area and find their way back to their base without the help of GPS signals.

Wildfires pose a significant natural disaster risk to populations and contribute to accelerated climate change. As wildfires are also affected by climate change, extreme wildfires are becoming increasingly frequent. Although they occur less frequently globally than those sparked by human activities, lightning-ignited wildfires play a substantial role in carbon emissions and account for the majority of burned areas in certain regions. While existing computational models, especially those based on machine learning, aim to predict lightning-ignited wildfires, they are typically tailored to specific regions with unique characteristics, limiting their global applicability. In this study, we present machine learning models designed to characterize and predict lightning-ignited wildfires on a global scale. Our approach involves classifying lightning-ignited versus anthropogenic wildfires, and estimating with high accuracy the probability of lightning to ignite a fire based on a wide spectrum of factors such as meteorological conditions and vegetation. Utilizing these models, we analyze seasonal and spatial trends in lightning-ignited wildfires shedding light on the impact of climate change on this phenomenon. We analyze the influence of various features on the models using eXplainable Artificial Intelligence (XAI) frameworks. Our findings highlight significant global differences between anthropogenic and lightning-ignited wildfires. Moreover, we demonstrate that, even over a short time span of less than a decade, climate changes have steadily increased the global risk of lightning-ignited wildfires. This distinction underscores the imperative need for dedicated predictive models and fire weather indices tailored specifically to each type of wildfire.

Large Language Model (LLM)-based recommendation systems provide more comprehensive recommendations than traditional systems by deeply analyzing content and user behavior. However, these systems often exhibit biases, favoring mainstream content while marginalizing non-traditional options due to skewed training data. This study investigates the intricate relationship between bias and LLM-based recommendation systems, with a focus on music, song, and book recommendations across diverse demographic and cultural groups. Through a comprehensive analysis conducted over different LLM-models, this paper evaluates the impact of bias on recommendation outcomes. Our findings reveal that bias is so deeply ingrained within these systems that even a simpler intervention like prompt engineering can significantly reduce bias, underscoring the pervasive nature of the issue. Moreover, factors like intersecting identities and contextual information, such as socioeconomic status, further amplify these biases, demonstrating the complexity and depth of the challenges faced in creating fair recommendations across different groups.

The request for fast response and safe operation after natural and man-made disasters in urban environments has spurred the development of robotic systems designed to assist in search and rescue operations within complex rubble sites. Traditional Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) face significant limitations in such confined and obstructed environments. This paper introduces a novel vine robot designed to navigate dense rubble, drawing inspiration from natural growth mechanisms found in plants. Unlike conventional robots, vine robots are soft robots that can grow by everting their material, allowing them to navigate through narrow spaces and obstacles. The prototype presented in this study incorporates pneumatic muscles for steering and oscillation, an equation-based robot length control plus feedback pressure regulating system for extending and retracting the robot body. We conducted a series of controlled experiments in an artificial rubble testbed to assess the robot performance under varying environmental conditions and robot parameters, including volume ratio, environmental weight, oscillation, and steering. The results show that the vine robot can achieve significant penetration depths in cluttered environments with mixed obstacle sizes and weights, and can maintain repeated trajectories, demonstrating potential for mapping and navigating complex underground paths. Our findings highlight the suitability of the vine robot for urban search and rescue missions, with further research planned to enhance its robustness and deployability in real-world scenarios.

This paper addresses the problem of autonomous UAV search missions, where a UAV must locate specific Entities of Interest (EOIs) within a time limit, based on brief descriptions in large, hazard-prone environments with keep-out zones. The UAV must perceive, reason, and make decisions with limited and uncertain information. We propose NEUSIS, a compositional neuro-symbolic system designed for interpretable UAV search and navigation in realistic scenarios. NEUSIS integrates neuro-symbolic visual perception, reasoning, and grounding (GRiD) to process raw sensory inputs, maintains a probabilistic world model for environment representation, and uses a hierarchical planning component (SNaC) for efficient path planning. Experimental results from simulated urban search missions using AirSim and Unreal Engine show that NEUSIS outperforms a state-of-the-art (SOTA) vision-language model and a SOTA search planning model in success rate, search efficiency, and 3D localization. These results demonstrate the effectiveness of our compositional neuro-symbolic approach in handling complex, real-world scenarios, making it a promising solution for autonomous UAV systems in search missions.

This paper pioneers the use of quantum machine learning (QML) for modeling the Ohmic contact process in GaN high-electron-mobility transistors (HEMTs) for the first time. Utilizing data from 159 devices and variational auto-encoder-based augmentation, we developed a quantum kernel-based regressor (QKR) with a 2-level ZZ-feature map. Benchmarking against six classical machine learning (CML) models, our QKR consistently demonstrated the lowest mean absolute error (MAE), mean squared error (MSE), and root mean squared error (RMSE). Repeated statistical analysis confirmed its robustness. Additionally, experiments verified an MAE of 0.314 ohm-mm, underscoring the QKR's superior performance and potential for semiconductor applications, and demonstrating significant advancements over traditional CML methods.

We are motivated primarily by the adaptation of text-to-speech synthesis models; however we argue that more generic parameter-efficient fine-tuning (PEFT) is an appropriate framework to do such adaptation. Nevertheless, catastrophic forgetting remains an issue with PEFT, damaging the pre-trained model's inherent capabilities. We demonstrate that existing Bayesian learning techniques can be applied to PEFT to prevent catastrophic forgetting as long as the parameter shift of the fine-tuned layers can be calculated differentiably. In a principled series of experiments on language modeling and speech synthesis tasks, we utilize established Laplace approximations, including diagonal and Kronecker-factored approaches, to regularize PEFT with the low-rank adaptation (LoRA) and compare their performance in pre-training knowledge preservation. Our results demonstrate that catastrophic forgetting can be overcome by our methods without degrading the fine-tuning performance, and using the Kronecker-factored approximation produces a better preservation of the pre-training knowledge than the diagonal ones.

However, most of the recent literature considers 3D motion data [4-6, 14, 42, 50, 53-55, 58], primarily because such data representation Co-speech gestures are fundamental for communication. The advent contains the depth dimension and is more easily transferable to of recent deep learning techniques has facilitated the creation downstream applications such as 3D virtual agents or social robots of lifelike, synchronous co-speech gestures for Embodied Conversational [43, 56]. But, it is not easy to collect high-quality 3D motion data, Agents. "In-the-wild" datasets, aggregating video content as one needs a motion capture setup in a controlled environment, from platforms like YouTube via human pose detection technologies, hence limiting the size and diversity of such datasets. To access provide a feasible solution by offering 2D skeletal sequences aligned 3D motion data and still gather large-scale datasets of diverse and with speech. Concurrent developments in lifting models enable the spontaneous gestures, multiple works leverage an estimation of the conversion of these 2D sequences into 3D gesture databases. However, 3D gestures inferred from 2D poses extracted from "in-the-wild" it is important to note that the 3D poses estimated from the videos [39, 50, 55]. Nevertheless, to convert extracted 2D keypoints 2D extracted poses are, in essence, approximations of the groundtruth, to 3D, one needs a third-party 2D-to-3D lifter, which may be prone which remains in the 2D domain. This distinction raises to inaccuracies, notably because of the ambiguous nature of 3D questions about the impact of gesture representation dimensionality pose estimation from 2D keypoints [44].

Learning with neural networks from a continuous stream of visual information presents several challenges due to the non-i.i.d. nature of the data. However, it also offers novel opportunities to develop representations that are consistent with the information flow. In this paper we investigate the case of unsupervised continual learning of pixel-wise features subject to multiple motion-induced constraints, therefore named motion-conjugated feature representations. Differently from existing approaches, motion is not a given signal (either ground-truth or estimated by external modules), but is the outcome of a progressive and autonomous learning process, occurring at various levels of the feature hierarchy. Multiple motion flows are estimated with neural networks and characterized by different levels of abstractions, spanning from traditional optical flow to other latent signals originating from higher-level features, hence called higher-order motions. Continuously learning to develop consistent multi-order flows and representations is prone to trivial solutions, which we counteract by introducing a self-supervised contrastive loss, spatially-aware and based on flow-induced similarity. We assess our model on photorealistic synthetic streams and real-world videos, comparing to pre-trained state-of-the art feature extractors (also based on Transformers) and to recent unsupervised learning models, significantly outperforming these alternatives.