arXiv preprints from January 1, 2026 through September 23, 2026 — 12:26:14 EST
A conversational AI that cannot tell its own output from what a user said will treat its own mistakes as user-provided facts. In humans, this capacity is called reality monitoring, and its failures are linked to hallucinations, delusions, and confabulation, yet whether LLMs possess it remains untested. Here we show, across two...
Mechanical instabilities provide a general design principle for shaping developing organs and engineering soft materials. However, in slender structures, simple elastic buckling tends to erase rather than preserve shape complexity: structures relax to the simplest possible shape, erasing finer detail. Living systems nonetheless build...
We are studying a new regulator of microtubule homeostasis, wdA, in the fungus Aspergillus nidulans. Mutations in wdA confer cold-sensitive lethal mitotic catastrophe, suggesting an important role in microtubule-based processes. Previous genetic studies showed that deletion of TBCA causes a near-lethal phenotype that is suppressed by...
Visual prostheses require users to interpret sparse and distorted artificial percepts through active visual search. We developed an EEG-guided neuroadaptive training platform for simulated prosthetic vision in immersive virtual reality and evaluated its feasibility in a sham-controlled object-localization task. Twenty-two sighted...
Identifying therapeutic target genes from single-cell RNA sequencing (scRNA-seq) data remains a fundamental challenge in translational biology. Unlike bulk assays, scRNA-seq captures heterogeneous cellular states and rare subpopulations, but this same heterogeneity makes target discovery highly sensitive to analytical choices...
This paper presents a robust mathematical framework for quantifying antiproliferative effects from noisy in vitro cell viability experiments. The methodology is demonstrated using crystal violet assay measurements of quinolinic acid-induced growth inhibition in B16-F10 murine melanoma, RAW264.7 macrophage, and HaCaT keratinocyte cells...
We consider the spatial spread of epidemic into an unstable, healthy state. When the transmission rate depends on the fraction of infected, the propagating pulse of infection can be either a pulled front or a pushed front. We determined the phase space of parameters for the pulled and pushed regions both by numerically solving the...
The rapid evolution of generative models has unlocked new potentials in protein binder design, a pivotal task in structural biology, by facilitating end-to-end generation via joint sequence-structure modeling or hallucination. However, existing approaches are predominantly implemented under a single-target, single-state assumption,...
Specific mixing or demixing of molecular species is a characteristic feature of condensed intracellular membraneless compartments. How sequence patterns of intrinsically disordered proteins (IDPs) fundamentally impact subcompartmentalization of biomolecular condensates and their role in buffering against concentration fluctuations are...
Gene regulatory network modeling often requires balancing predictive accuracy and mechanistic interpretability. In this work, we compare continuous surrogate models and a discrete mechanistic model on the same \textit{Arabidopsis thaliana} induced systemic resistance (ISR) dataset, using both the raw continuous gene-expression...
Learning causal graphs from interventional data is a challenging problem with broad applications. In molecular biology, for example, a central goal is to uncover gene regulatory networks from large-scale perturbation data. An ideal algorithm for this task should scale to thousands of nodes, incorporate interventions even when their...
We present Loom, a spatial transcriptomics (ST) visual computing system to support the analysis of pseudo-temporal trajectories, comparative investigation across samples and regions of interest, and the examination of spatially structured processes within local microenvironments. ST is a molecular profiling technology that measures...
Multimodal foundation models learned to see and to speak by consuming the whole internet. Embodied agents admit no such shortcut, since they require data that couple observations with physical states and actions. These signals can be provided, to varying degrees, by multiple data sources. In this work, we organize the embodied data...
Multimodal large language models (MLLMs) hold immense potential to revolutionize clinical practice, yet deploying them in the medical domain is fundamentally a vision-centric challenge: models must absorb knowledge from heterogeneous 2D and 3D medical images, and evaluation protocols must align with radiologists' clinical practice and...
Neutrino oscillations establish that neutrinos are massive, providing the only laboratory detection of physics beyond the Standard Model. Direct kinematic experiments bound the electron-neutrino mass to $m_{ν_e} < 0.45$ eV (KATRIN, 90% CL), implying $\sum m_ν\lesssim 1.3$ eV. Conversely, cosmology within $Λ$CDM is highly constraining:...
Dynamic applications, including optimal-transport Flow Matching, repeatedly solve related entropic optimal transport problems, yet conventional distributed Sinkhorn processes frames sequentially and synchronizes after every iteration. We present TemporalSinkhorn, a parallel-in-time executor that batches future candidates and their...
Although anyon condensation is a standard mechanism for relating topological orders, anyon condensation in symmetry-enriched topological (SET) phases is more intricate because the condensate must also be compatible with the global symmetry. We study symmetry-preserving anyon condensation in SET phases described by the enlarged...
Here we analyze the creation of quantum simulacra: phenomena that emerge from treating a Hermitian or non-Hermitian quantum system in metrics other than the standard $L^{2}$. Changing the metric redefines the set of system observables and thus the experimental arrangement for their measurement, making quantum contextuality and...
An antidirected path is an oriented path in which every vertex sees either just incoming or just outgoing edges. We prove that every oriented graph with minimum semidegree at least $k$ contains an antidirected path of length $2 k -1$. This confirms a conjecture of Stein.
High-contrast imaging of Solar System scale exoplanets and protoplanets demands advancements in instrumentation to access deeper starlight suppression at smaller angular separations than today's state-of-the-art. The multi-mode to single-mode conversion capabilities of photonic lanterns (PLs) provide new avenues to implement...
The use of cheat sheets in exams is often framed as a way to reduce cognitive load and support student performance. However, little is known about how students choose between self-created and instructor-provided cheat sheets, or how these choices relate to their broader approaches to exam preparation. We conducted a longitudinal study...
We prove that area-minimizing submanifolds in mod $2$ homology are not generically smooth, except in the case of geodesics, minimal surfaces and minimal hypersurfaces. This settles a conjecture of White that asks the generic smoothness of area-minimizing submanifolds in mod $2$ homology. We furthermore establish a lower bound on the...
We consider a cosmological first-order phase transition in which bubbles of true vacuum nucleate during inflation but do not collide and percolate until the Universe has entered the radiation-dominated era. If the collisions take place soon after the end of inflation, the size of these bubbles can be significantly greater than the...
Thermally driven atmospheric escape, including photo-evaporation and core-powered mass-loss, plays a key role in shaping the evolution of close-in exoplanets, yet most current models rely on simplified one-dimensional descriptions of atmospheric escape. In this work, we perform 3D hydrodynamic simulations of atmospheric outflows from...
Motivated by learning from heterogeneous and overlapping data providers, we study a stylized model of distribution learning from restricted conditional samples. The goal is to learn an unknown distribution $p$ on a finite domain $[n]$. The learner is given a fixed family of queryable sets $\mathscr{S} \subseteq 2^{[n]}$, and each...