Umar Faruk Abubacar

Tiny Recursive Models (TRM) achieve strong results on reasoning tasks through iterative refinement of a shared network. We investigate whether these recursive mechanisms transfer to Quality Estimation (QE) for low-resource languages using a three-phase methodology. Experiments on 8 language pairs on a low-resource QE dataset reveal three findings. First, TRM’s recursive mechanisms do not transfer to QE. External iteration hurts performance, and internal recursion offers only narrow benefits. Next, representation quality dominates architectural choices, and lastly, frozen pretrained embeddings match fine-tuned performance while reducing trainable parameters by 37× (7M vs 262M). TRM-QE with frozen XLM-R embeddings achieves a Spearman’s correlation of 0.370, matching fine-tuned variants (0.369) and outperforming an equivalent-depth standard transformer (0.336). On Hindi and Tamil, frozen TRM-QE outperforms MonoTransQuest (560M parameters) with 80× fewer trainable parameters, suggesting that weight sharing combined with frozen embeddings enables parameter efficiency for QE.

Pseudomonas aeruginosa is a prevalent healthcare pathogen. It forms biofilms, which significantly increase antibiotic resistance, escalating the burden on healthcare systems. While global biofilm phenomena are observable experimentally, understanding local behaviours of bacteria and their environment, and how these factors drive biofilm formation, is challenging. Agent-based modelling (ABM) offers a computational solution to simulate these local behaviours and resulting large-scale phenomena. ABMs can simulate physical mechanisms like erosion and sloughing but often lack the ability to capture them during biofilm growth and validate them experimentally. We present a three-dimensional ABM of P. aeruginosa biofilms growing under constant flow. This model is the first of its kind to accurately capture the change in physical shape of biofilms growing under constant flow, as well as capture erosion and sloughing as natural byproducts of local bacteria interactions. The model is validated both qualitatively and quantitatively against experimental culture results, providing a general framework that can be adapted to simulate different environments or other biofilm-forming species.

Neuroscientists are increasingly initiating large-scale collaborations which bring together tens to hundreds of researchers. At this scale, such projects can tackle big challenges and engage diverse participants. Inspired by projects in mathematics, we set out to test the feasibility of widening access to such projects even further, by running a massively collaborative project in computational neuroscience. The key difference, with prior neuroscientific efforts, being that our entire project (code, results, and writing) was public from the outset, and that anyone could participate. To achieve this, we launched a public Git repository, with code for training spiking neural networks to solve a sound localization task via surrogate gradient descent. We then invited anyone, anywhere to use this code as a springboard for exploring questions of interest to them, and encouraged participants to share their work both asynchronously through Git and synchronously at online workshops. Our hope was that the resulting range of participants would allow us to make discoveries that a single team would have been unlikely to find. At a scientific level, our work investigated how a range of biological parameters, from time delays to membrane time constants and levels of inhibition, could impact sound localization in networks of spiking units. At a more macro-level, our project brought together researchers from multiple countries, provided hands-on research experience to early career participants and opportunities for supervision and teaching to later career participants. While our scientific results were not groundbreaking, our project demonstrates the potential for massively collaborative projects to transform neuroscience.

Neuroscientists are increasingly initiating large-scale collaborations which bring together tens to hundreds of researchers. However, while these projects represent a step-change in scale, they retain a traditional structure with centralised funding, participating laboratories and data sharing on publication. Inspired by an open-source project in pure mathematics, we set out to test the feasibility of an alternative structure by running a grassroots, massively collaborative project in computational neuroscience. To do so, we launched a public Git repository, with code for training spiking neural networks to solve a sound localisation task via surrogate gradient descent. We then invited anyone, anywhere to use this code as a springboard for exploring questions of interest to them, and encouraged participants to share their work both asynchro-nously through Git and synchronously at monthly online workshops. At a scientific level, our work investigated how a range of biologically-relevant parameters, from time delays to mem-brane time constants and levels of inhibition, could impact sound localisation in networks of spiking units. At a more macro-level, our project brought together 31 researchers from multiple countries, provided hands-on research experience to early career participants, and opportunities for supervision and teaching to later career participants. Looking ahead, our project provides a glimpse of what open, collaborative science could look like and provides a necessary, tentative step towards it.