Multiperspective Fusion (MPF) is a novel posttraining alignment framework for large language models (LLMs) developed in response to the growing need for easy bias mitigation. Built on top of the SAGED pipeline, an automated system for constructing bias benchmarks and extracting interpretable baseline distributions, MPF leverages multiperspective generations to expose and align biases in LLM outputs with nuanced, humanlike baselines. By decomposing baseline, such as sentiment distributions from HR professionals, into interpretable perspective components, MPF guides generation through sampling and balancing of responses, weighted by the probabilities obtained in the decomposition. Empirically, we demonstrate its ability to align LLM sentiment distributions with both counterfactual baselines (absolute equality) and the HR baseline (biased for Top Univeristy), resulting in small KL divergence, reduction of calibration error and generalization to unseen questions. This shows that MPF offers a scalable and interpretable method for alignment and bias mitigation, compatible with deployed LLMs and requiring no extensive prompt engineering or finetuning.

Recently, innovative adaptations of the Ritz Method incorporating deep learning have been developed, known as the Deep Ritz Method. This approach employs a neural network as the test function for variational problems. However, the neural network does not inherently satisfy the boundary conditions of the variational problem. To resolve this issue, the Deep Ritz Method introduces a penalty term into the functional of the variational problem, which can lead to misleading results during the optimization process. In this work, an ansatz is proposed that inherently satisfies the boundary conditions of the variational problem. The results demonstrate that the proposed ansatz not only eliminates misleading outcomes but also reduces complexity while maintaining accuracy, showcasing its practical effectiveness in addressing variational problems.

Accurate and robust localization plays a key role for autonomous aerial robots. Today, LiDAR-based and camerabased (indoors and outdoors) and GNSS-based (outdoors) solutions are widely used. However, the emergence of flapping-wing robots [1-3] has motivated a paradigm change. First, the limited payload and the resource-constrained computation impose a limitation on the number and type of sensors to be mounted [4]. Second, ornithopters' flapping strokes entail several challenges for perception (e.g., motion blur in cameras) [5]. Those restrictions are even more critical in the case of flapping-wing micro air vehicles (FWMAV) [6].

Abstract--This work presents a prototype of a multirotor aerial vehicle capable of precision landing, even under the effects of rotor failures. The manuscript presents the fault-tolerant techniques and mechanical designs to achieve a fault-tolerant multirotor, and a vision-based navigation system required to achieve a precision landing. Preliminary experimental results will be shown, to validate on one hand the fault-tolerant control vehicle and, on the other hand, the autonomous landing algorithm. Also, a prototype of the fault-tolerant UAV is presented, capable of precise autonomous landing, which will be used in future experiments. On the bottom-left arm, a servo allows to tilt the re-configurable motor.

However, Spanish is occasionally adopted as the focus language for research endeavors and as result multiple projects are conducted in Spanish to explore language-specific nuances and challenges in NLP applications. Named-Entity recognition [1], Machine Translation [2], Semantic Relation Extraction [3] among others tasks have been conducted with a focus on Spanish language data, allowing for a more nuanced understanding of the intricacies involved. In this paper we present Classifier for Named Entities Recognized (CNER) a linguistically-aware online service that offers the possibility to test two main tasks of NLP, Named Entity Recognition (NER) and Relation Extraction (RE) for Spanish language. This together with other projects on Spanish language have been evaluated and adapted as a web service. In this context, language technologies and natural language processing (NLP) tools can support the identification of useful information in text and to promote its understanding. Specifically, CNER i) identifies the mentions follow the ACE standard with entity types include Person (PER), Organisation (ORG), Facility (FAC), Location (LOC), Geographical/Political (GPE), Vehicle (VEH), Vehicle (VEH) and Weapon (WEA) [4], [5]; ii) displays three different NER tools as previous step to RE task and iii) offers entity relationship information through tags GPE-AFF, PHYS, DISC, EMP-ORG, ART, NON-REL representing the relations between two entities [6] .

Support Vector Machine (SVM) is a powerful tool in binary classification, known to attain excellent misclassification rates. On the other hand, many realworld classification problems, such as those found in medical diagnosis, churn or fraud prediction, involve misclassification costs which may be different in the different classes. However, it may be hard for the user to provide precise values for such misclassification costs, whereas it may be much easier to identify acceptable misclassification rates values. In this paper we propose a novel SVM model in which misclassification costs are considered by incorporating performance constraints in the problem formulation. Specifically, our aim is to seek the hyperplane with maximal margin yielding misclassification rates below given threshold values. Such maximal margin hyperplane is obtained by solving a quadratic convex problem with linear constraints and integer variables. The reported numerical experience shows that our model gives the user control on the misclassification rates in one class (possibly at the expense of an increase in misclassification rates for the other class) and is feasible in terms of running times.

Vision Transformers (ViTs) have successfully been applied to image classification problems where large annotated datasets are available. On the other hand, when fewer annotations are available, such as in biomedical applications, image augmentation techniques like introducing image variations or combinations have been proposed. However, regarding ViT patch sampling, less has been explored outside grid-based strategies. In this work, we propose Random Vision Transformer Tokens (RaViTT), a random patch sampling strategy that can be incorporated into existing ViTs. We experimentally evaluated RaViTT for image classification, comparing it with a baseline ViT and state-of-the-art (SOTA) augmentation techniques in 4 datasets, including ImageNet-1k and CIFAR-100. Results show that RaViTT increases the accuracy of the baseline in all datasets and outperforms the SOTA augmentation techniques in 3 out of 4 datasets by a significant margin +1.23% to +4.32%. Interestingly, RaViTT accuracy improvements can be achieved even with fewer tokens, thus reducing the computational load of any ViT model for a given accuracy value.

One of the motivations for exploring flapping-wing aerial robotic systems is to seek energy reduction, by maintaining manoeuvrability, compared to conventional unmanned aerial systems. A Flapping Wing Flying Robot (FWFR) can glide in favourable wind conditions, decreasing energy consumption significantly. In addition, it is also necessary to investigate the power consumption of the components in the flapping-wing robot. In this work, two sets of the FWFR components are analyzed in terms of power consumption: a) motor/electronics components and b) a vision system for monitoring the environment during the flight. A measurement device is used to record the power utilization of the motors in the launching and ascending phases of the flight and also in cruising flight around the desired height. Additionally, an analysis of event cameras and stereo vision systems in terms of energy consumption has been performed. The results provide a first step towards decreasing battery usage and, consequently, providing additional flight time.

The Department of Geosciences at the Universidad de Los Andes in Bogotá, Colombia, invites applications for one tenure-track faculty position in Geophysics, with particular emphasis in the areas of Seismology, Seismic Hazards or Seismic Exploration. We encourage candidates whose research integrates numerical modeling or Artificial Intelligence techniques with data analysis and processing. Applicants must hold a Ph.D. degree, ideally a relevant postdoctoral experience, and should have a significant record of research experience documented by peer-reviewed publications. Candidates with relevant experience in seismological observatories/networks or industry are also encouraged to apply. Fluency in Spanish language is preferred but not compulsory.

Event cameras can capture pixel-level illumination changes with very high temporal resolution and dynamic range. They have received increasing research interest due to their robustness to lighting conditions and motion blur. Two main approaches exist in the literature to feed the event-based processing algorithms: packaging the triggered events in event packages and sending them one-by-one as single events. These approaches suffer limitations from either processing overflow or lack of responsivity. Processing overflow is caused by high event generation rates when the algorithm cannot process all the events in real-time. Conversely, lack of responsivity happens in cases of low event generation rates when the event packages are sent at too low frequencies. This paper presents ASAP, an adaptive scheme to manage the event stream through variable-size packages that accommodate to the event package processing times. The experimental results show that ASAP is capable of feeding an asynchronous event-by-event clustering algorithm in a responsive and efficient manner and at the same time prevents overflow.