arXiv subjects

Craig Barton

Publications and source records attributed to Craig Barton.

11 recordsLinked to original sources

arXiv2025

AI tutoring can safely and effectively support students: An exploratory RCT in UK classrooms

One-to-one tutoring is widely considered the gold standard for personalized education, yet it remains prohibitively expensive to scale. To evaluate whether generative AI might help expand access to this resource, we conducted an exploratory randomized controlled trial (RCT) with $N = 165$ students across five UK secondary schools. We integrated LearnLM -- a generative AI model fine-tuned for pedagogy -- into chat-based tutoring sessions on the Eedi mathematics platform. In the RCT, expert tutors directly supervised LearnLM, with the remit to revise each message it drafted until they would be satisfied sending it themselves. LearnLM proved to be a reliable source of pedagogical instruction, with supervising tutors approving 76.4% of its drafted messages making zero or minimal edits (i.e., changing only one or two characters). This translated into effective tutoring support: students guided by LearnLM performed at least as well as students chatting with human tutors on each learning outcome we measured. In fact, students who received support from LearnLM were 5.5 percentage points more likely to solve novel problems on subsequent topics (with a success rate of 66.2%) than those who received tutoring from human tutors alone (rate of 60.7%). In interviews, tutors highlighted LearnLM's strength at drafting Socratic questions that encouraged deeper reflection from students, with multiple tutors even reporting that they learned new pedagogical practices from the model. Overall, our results suggest that pedagogically fine-tuned AI tutoring systems may play a promising role in delivering effective, individualized learning support at scale.

cs.CY↗

arXiv2025

Thermoelectric Fingerprinting of Bloch- and N\'{e}el-type Skyrmions

Magnetic skyrmions are nanoscale spin textures that exhibit topological stability, which, along with novel thermal and electrical transport properties, make them the ideal candidates for a variety of novel technological applications. Accessing the skyrmion spin texture at the nanoscale and understanding its interaction with local thermal gradients is essential for engineering skyrmion-based transport phenomena. However, direct experimental insight into the local thermoelectric response of single skyrmions remains limited. To address this, we employ scanning thermoelectric microscopy~(SThEM) to probe the nanoscale thermoelectric response from a single skyrmion. By mapping the local thermoelectric voltage with nanoscale precision, we reveal a unique spatially resolved response that is the convolution of the underlying spin texture of the skyrmion and its interaction with the highly localised thermal gradient originating from the heated probe. We combine this with thermoelectric modelling of a range of skyrmion spin textures to reveal unique thermoelectric responses and allow the possibility of SThEM to be used as a tool to distinguish nanoscale spin textures. These findings provide fundamental insights into the interaction of topologically protected spin textures with local thermal gradients and the resultant spin transport. We demonstrate a novel route to characterise nanoscale spin textures, accelerating the material optimisation cycle, while also opening the possibility to harness skyrmions for spin caloritronics.

cond-mat.mes-hall↗

arXiv2025

Skyrmion motion in a synthetic antiferromagnet driven by asymmetric spin wave emission

Skyrmions have been proposed as new information carriers in racetrack memory devices. To realise such devices, a small size; high speed of propagation; and minimal skyrmion Hall angle are required. Synthetic antiferromagnets (SAFs) present the ideal materials system to realise these aims. In this work, we use micromagnetic simulations to propose a new method for manipulating them using exclusively global magnetic fields. An out-of-plane microwave field induces oscillations in the skyrmions radius which in turn emits spin waves. When a static in-plane field is added, this breaks the symmetry of the skyrmions and causes asymmetric spin wave emission. This in turn drives motion of the skyrmions, with the fastest velocities observed at the frequency of the intrinsic out-of-phase breathing mode of the pair of skyrmions. This behaviour is investigated over a range of experimentally realistic antiferromagnetic interlayer exchange coupling strengths, and the results compared to previous works studying similar motion driven with an oscillating electric field. Through this the true effect of varying the exchange coupling strength is determined, and greater insight is gained into the mechanism of skyrmion motion. These results will help to inform the design of future novel computing architectures based on the dynamics of skyrmions in synthetic antiferromagnets.

cond-mat.mes-hall↗

arXiv2024

Phase Coexistence and Transitions between Antiferromagnetic and Ferromagnetic States in a Synthetic Antiferromagnet

In synthetic antiferromagnets (SAFs) the combination of antiferromagnetic order and synthesis using conventional sputtering techniques is combined to produce systems that are advantageous for spintronics applications. Here we present the preparation and study of SAF multilayers possessing both perpendicular magnetic anisotropy and the Dzyaloshinskii-Moriya interaction. The multilayers have an antiferromagnetically (AF) aligned ground state but can be forced into a full ferromagnetic (FM) alignment by applying an out-of-plane field $\sim 100$~mT. We study the spin textures in these multilayers in their ground state as well as around the transition point between the AF and FM states, at fields $\sim 40$~mT, by imaging the spin textures using complementary methods: photo-emission electron, magnetic force, and Lorentz transmission electron microscopies. The transformation into a FM state by field proceeds by a nucleation and growth process, where first skyrmionic nuclei form, which broaden into regions containing a FM-aligned labyrinth pattern that eventually occupies the whole film. This process remarkably occurs without any significant change in the net magnetic moment of the multilayer. The mix of AF- and FM-aligned regions on the micron scale in the middle of this transition is reminiscent of a first-order phase transition that exhibits phase coexistence. These results are important for guiding the design of spintronic devices using chiral magnetic textures made from SAFs.

cond-mat.mes-hall↗

arXiv2022

NeurIPS Competition Instructions and Guide: Causal Insights for Learning Paths in Education

In this competition, participants will address two fundamental causal challenges in machine learning in the context of education using time-series data. The first is to identify the causal relationships between different constructs, where a construct is defined as the smallest element of learning. The second challenge is to predict the impact of learning one construct on the ability to answer questions on other constructs. Addressing these challenges will enable optimisation of students' knowledge acquisition, which can be deployed in a real edtech solution impacting millions of students. Participants will run these tasks in an idealised environment with synthetic data and a real-world scenario with evaluation data collected from a series of A/B tests.

cs.CY↗

arXiv2021

Results and Insights from Diagnostic Questions: The NeurIPS 2020 Education Challenge

This competition concerns educational diagnostic questions, which are pedagogically effective, multiple-choice questions (MCQs) whose distractors embody misconceptions. With a large and ever-increasing number of such questions, it becomes overwhelming for teachers to know which questions are the best ones to use for their students. We thus seek to answer the following question: how can we use data on hundreds of millions of answers to MCQs to drive automatic personalized learning in large-scale learning scenarios where manual personalization is infeasible? Success in using MCQ data at scale helps build more intelligent, personalized learning platforms that ultimately improve the quality of education en masse. To this end, we introduce a new, large-scale, real-world dataset and formulate 4 data mining tasks on MCQs that mimic real learning scenarios and target various aspects of the above question in a competition setting at NeurIPS 2020. We report on our NeurIPS competition in which nearly 400 teams submitted approximately 4000 submissions, with encouragingly diverse and effective approaches to each of our tasks.

cs.CY↗

arXiv2020

Instructions and Guide for Diagnostic Questions: The NeurIPS 2020 Education Challenge

Digital technologies are becoming increasingly prevalent in education, enabling personalized, high quality education resources to be accessible by students across the world. Importantly, among these resources are diagnostic questions: the answers that the students give to these questions reveal key information about the specific nature of misconceptions that the students may hold. Analyzing the massive quantities of data stemming from students' interactions with these diagnostic questions can help us more accurately understand the students' learning status and thus allow us to automate learning curriculum recommendations. In this competition, participants will focus on the students' answer records to these multiple-choice diagnostic questions, with the aim of 1) accurately predicting which answers the students provide; 2) accurately predicting which questions have high quality; and 3) determining a personalized sequence of questions for each student that best predicts the student's answers. These tasks closely mimic the goals of a real-world educational platform and are highly representative of the educational challenges faced today. We provide over 20 million examples of students' answers to mathematics questions from Eedi, a leading educational platform which thousands of students interact with daily around the globe. Participants to this competition have a chance to make a lasting, real-world impact on the quality of personalized education for millions of students across the world.

cs.CY↗

arXiv2020

Local spin Seebeck imaging with scanning thermal probe

In this work we present the results of an experiment to locally resolve the spin Seebeck effect in a high-quality Pt/YIG sample. We achieve this by employing a locally heated scanning thermal probe to generate a highly local non-equilibrium spin current. To support our experimental results, we also present a model based on the non-equilibrium thermodynamic approach which is in a good agreement with experimental findings. To further corroborate our results, we index the locally resolved spin Seebeck effect with that of the local magnetisation texture by MFM and correlate corresponding regions. We hypothesise that this technique allows imaging of magnetisation textures within the magnon diffusion length and hence characterisation of spin caloritronic materials at the nanoscale.

cond-mat.mes-hall↗

arXiv2020

Thermoelectric signature of individual skyrmions

We experimentally study the thermoelectrical signature of individual skyrmions in chiral Pt/Co/Ru multilayers. Using a combination of controlled nucleation, single skyrmion annihilation, and magnetic field dependent measurements the thermoelectric signature of individual skyrmions is characterized. The observed signature is explained by the anomalous Nernst effect of the skyrmions spin structure. Possible topological contributions to the observed thermoelectrical signature are discussed. Such thermoelectrical characterization allows for non-invasive detection and counting of skyrmions and enables fundamental studies of topological thermoelectric effects on the nano scale

cond-mat.mes-hall↗

arXiv2019

Diameter-independent skyrmion Hall angle in the plastic flow regime observed in chiral magnetic multilayers

Magnetic skyrmions are topologically non-trivial nanoscale objects. Their topology, which originates in their chiral domain wall winding, governs their unique response to a motion-inducing force. When subjected to an electrical current, the chiral winding of the spin texture leads to a deflection of the skyrmion trajectory, characterized by an angle with respect to the applied force direction. This skyrmion Hall angle was believed to be skyrmion diameter-dependent. In contrast, our experimental study finds that within the plastic flow regime the skyrmion Hall angle is diameter-independent. At an average velocity of 6 $\pm$ 1 m/s the average skyrmion Hall angle was measured to be 9{\deg} $\pm$ 2{\deg}. In fact, in the plastic flow regime, the skyrmion dynamics is dominated by the local energy landscape such as materials defects and the local magnetic configuration.

cond-mat.mes-hall↗

arXiv2019

European research on magnetic nanoparticles for biomedical applications: standardisation aspects

Magnetic nanoparticles have many applications in biomedicine and other technical areas. Despite their huge economic impact, there are no standardised procedures available to measure their basic magnetic properties. The International Organization for Standardization is working on a series of documents on the definition of characteristics of magnetic nanomaterials. We review previous and ongoing European research projects on characteristics of magnetic nanoparticles and present results of an online survey among European researchers.

physics.med-ph↗