Frontier Research Intelligence
Preparing the latest research view.
Browse
Search and filter the research library by collection and research area. Results are URL-driven for shareable views.
Recent works have highlighted the use of dual nostril 'stereo' olfaction by a variety of animals. In this work we perform "back of the envelope" calculations to determine when stereo olfaction is useful compared to single nostril 'mono' olfaction. We find that stereo olfaction is advantageous when there are large relative changes in the odour concentration, and when the spatial length scales of correlations in air...
Quantum machine learning (QML) algorithms in high energy physics (HEP) can efficiently represent and leverage long range, high order correlations in high dimensional collider data, potentially with fewer parameters and favorable scaling relative to classical models. Deployment of QML in real time collider applications such as trigger systems requires the ability to emulate and compile quantum circuits classically,...
Correlation alignment and the maximum mean discrepancy are two widely used distribution matching frameworks for unsupervised domain adaptation (UDA). However, high variance in these losses has been shown to undermine their effectiveness in minibatch optimisation settings. Furthermore, the losses lack finite sum structure, which renders them incompatible with classical stochastic variance reduction (SVR) methods. T...
This repository includes the appendix to the following submission: "Neurotechnology and Privacy: Understanding User Perspectives on Brain Computer Interface Data Collection and Use", V. Susewind & S. Feldmann & E. Kablo & P. Arias Cabarcos. 2026. Structure: Survey instrument Patricipants' demographics and background tables
Natural language autoencoders score explanations of hidden activations by reconstruction: an explanation is deemed faithful if the activation can be regenerated from it. The test is structurally insensitive to individual false claims: if flipping a claim does not change the reconstruction, the claim is never penalized. We show the test is passed in two ways, neither faithful. On a released Qwen 2.5 7B verbalizer, ...
Full sized humanoid robot capabilities have grown exponentially in recent years, aiming towards general purpose deployment in human environments. A popular control method used by manufacturers utilizes Virtual Reality for upper body teleoperation and Reinforcement Learning for lower body balance and locomotion control. As a result, a single remote operator can see, manipulate, and navigate about a real, distant ph...
With growing surgeon shortages, automating surgical sub tasks such as tissue dissection offers a promising step toward reducing workload and expanding patient access. Prior work has relied on hand crafted incision policies that cannot quantify uncertainty or has relied on simulation based methods that require strong modeling assumptions. We instead view tissue attachment identification as an inherently probabilist...
Circuit analysis can support not only model explanation but also downstream interventions such as pruning, editing, steering, and selective fine tuning. However, conducting such analyses currently requires stitching together separate implementations for discovery, evaluation, and intervention, as well as hand authoring the contrastive prompts required by many discovery methods. This fragmentation makes methods dif...
As AI agents begin to automate AI R&D, we need ways to assess whether their outputs are safe to deploy, even when the agents themselves may be untrusted. AI control offers one such approach: rather than trusting the agent, it treats it as a potential adversary and uses a monitor to detect covert sabotage before deployment. We evaluate AI control for automated AI R&D with ResearchArena, a framework spanning four lo...
Autonomous flight in cluttered environments requires a robot to build a geometric map of its surroundings and plan safe, dynamically feasible trajectories, all onboard and in real time. Conventional approaches treat mapping and planning as separate stages and often rely on binary occupancy for collision checking. We argue that these two stages should be co designed around a single representation: a signed distance...
As LLM adoption becomes more widespread, there is a growing interest in detecting LLM generated content, for example through LLM detection tools and through heuristics based on language patterns. Detectors operate as an intervention that steers not only the detected attribute itself, but also downstream metrics such as LLM usage and output quality. In this work, we demonstrate how imperfect LLM detectors lead to c...
Reinforcement learning with verifiable rewards (RLVR) is rapidly advancing the reasoning capabilities of language models, yet the optimization layer that converts reward feedback into weight space updates remains poorly understood. Building on our prior analysis (Zhu et al., 2025), we study this missing layer through the singular structure of model weights and identify spectral inheritance: RLVR can reuse the base...
Reinforcement learning with verifiable rewards (RLVR) improves reasoning in large language models. Yet, typical RLVR approaches fail on difficult problems: when a model cannot generate any correct solutions, it receives \textit{zero} learning signal. Providing privileged guidance during training, such as solution prefixes, can help overcome this learning cliff by steering the model towards {correct solutions with ...
Coding agents increasingly operate in executable environments where a failed attempt produces actionable feedback rather than merely an incorrect answer. Existing cost aware systems typically treat such failures as cascade decisions: try a cheap model first, then escalate hard cases to a stronger and more expensive model. In coding, however, execution feedback can also make further cheap model recovery worthwhile,...
Speculative decoding, in which a lightweight draft model first generates a draft sequence that is then verified in parallel by the target model, has become a prevalent paradigm for accelerating large language model inference. Recent work such as DFlash further boosts drafting efficiency by leveraging diffusion drafters, whose parallel denoising mechanism enables draft generation in a single forward pass. In this w...
Motivated by the numerical computation of the Mean Escape Time (MET) $τ:Ω\to\mathbb{R}$ of a stochastic process from a bounded domain $Ω\subseteq\mathbb{R}^d$, we study elliptic Dirichlet boundary value problems (BVPs) using boundary enforced Physics Informed Neural Networks (PINNs), in which the Dirichlet condition is imposed exactly by multiplying the network output with a predefined distance to boundary approxi...
Stroke is a main reason for long term disability globally, motor and cognitive impairments appear after stroke, which can reduce quality of life for patients and makes the rehabilitation process become more complex, new auxiliary tools are needed. Non invasive brain computer interfaces (BCIs) are promising tools to help stroke rehabilitation. BCI combined with functional electrical stimulation (BCI FES) is used in...
Advances in neural recording, real time decoding and bioelectronic stimulation have enabled a new class of systems that re establish functional communication across disrupted neural pathways. In the context of paralysis, these neural bypass interfaces link upstream neural intent to effector activation downstream of the lesion, effectively circumventing sites of injury within the nervous system to restore volitiona...
Designing small molecule ligands that bind with high affinity to specific protein pockets is a fundamental goal in drug discovery, as small molecules constitute a major fraction of approved therapeutics. Recent breakthroughs in structure prediction, such as AlphaFold 3 and Boltz 2, enable accurate biomolecular interaction prediction and show promise as foundation models for downstream tasks, including binding affi...
The report envisions a decade in which drones move goods, medical supplies, and information at a scale comparable to national infrastructure investments like highways and the electric grid. Potential applications include natural disaster detection drones that spot wildfire sources within minutes, medical supply chains that bypass ground congestion to reach rural hospitals, and nationwide fleets that continuously i...
