Search for: All records

computation -- and schedules it on an independent pool of resources. Existing systems co-locate core attention with other components. At long context, the quadratic growth of CA computation and near-linear growth of the rest create load imbalance -- hence stragglers across data and pipeline groups. CAD is enabled by two key observations: (i) \emph{statelessness}: CA has no trainable parameters and minimal transient state, so balancing reduces to scheduling compute-bound tasks; and (ii) \emph{composability}: modern attention kernels sustain high utilization on fused batches of arbitrary-length token-level shards. CAD dynamically partitions the core attention computation into token-level tasks (CA-tasks), and dispatches them to a pool of devices specialized for CA computation (attention servers). It then rebatches CA-tasks to equalize CA compute across attention servers without loss of kernel efficiency. We have implemented CAD in a system called DistCA with a ping-pong scheme to completely overlap communication with compute, and in-place attention servers to improve memory utilization. Scaling to 512 H200 GPUs and 512K context length, DistCA eliminates DP/PP stragglers, achieves near-perfect compute and memory balance, and improves end-to-end training throughput by up to 1.9× over Megatron-LM and 1.35× over existing load-balancing methods. 

Cis-regulatory elements are specific DNA sequences that control gene expression in a spatiotemporal manner, and variation within these elements represents a major source of phenotypic diversity and evolutionary innovation. Nevertheless, how regulatory elements evolve and shape gene expression remains poorly understood, particularly in plants. The well-resolved phylogeny of allopolyploid peanut (Arachis hypogaea) and its diploid progenitors,Arachis duranensisandArachis ipaensis, provides an ideal system to investigate the regulatory evolution at a lineage-specific level. By integrating comparative analyses of sequence similarity, chromatin accessibility, histone modifications, conserved noncoding sequences (CNSs), and gene expression, we reconstructed the evolutionary trajectories of Accessible Chromatin Regions (ACRs), where regulatory elements typically reside, and revealed their distinct contributions to homoeolog expression bias, unequal expressions between homoeologs. Most ACRs exhibited high sequence similarity, comparable chromatin accessibility, and conserved states for H3K4me3, H3K56ac, and H3K36me3, indicating regulatory stability after hybridization and polyploidization. However, a subset of novel ACRs emerged de novo from previously nonregulatory regions or through sequence mutations in preexisting ACRs, arising at different rates and evolutionary stages. Notably, even highly sequence-conserved ACRs exhibited substantial variation in chromatin accessibility, consistent with CNS composition differences and minor sequence variation, although causal relationships remain to be demonstrated. Our analyses further revealed a complex spectrum of CNS dynamics within the diploid-polyploid framework. Overall, our study provides empirical insights into the fine-scale evolution of plant regulatory landscapes and complements previous large-scale comparisons across distant lineages. 

Abstract Training of neural networks (NNs) has emerged as a major consumer of both computational and energy resources. Quantum computers were coined as a root to facilitate training, but no experimental evidence has been presented so far. Here we demonstrate that quantum annealing platforms, such as D-Wave, can enable fast and efficient training of classical NNs, which are then deployable on conventional hardware. From a physics perspective, NN training can be viewed as a dynamical phase transition: the system evolves from an initial spin glass state to a highly ordered, trained state. This process involves eliminating numerous undesired minima in its energy landscape. The advantage of annealing devices is their ability to rapidly find multiple deep states. We found that this quantum training achieves superior performance scaling compared to classical backpropagation methods, with a clearly higher scaling exponent (1.01 vs. 0.78). It may be further increased up to a factor of 2 with a fully coherent quantum platform using a variant of the Grover algorithm. Furthermore, we argue that even a modestly sized annealer can be beneficial to train a deep NN by being applied sequentially to a few layers at a time. 

Abstract Understanding the mechanisms underlying key agricultural traits remains a central challenge in crop research, but recent advances in technologies are providing powerful tools to address this issue. Among these, single-cell and spatial transcriptomics have revealed tissue heterogeneity and spatial organization, offering unique insights into cellular gene expression dynamics and the coordinated activity of multiple cell types. These approaches help uncover how specific cell types contribute to agricultural traits and refine candidate loci lists through integration with trait-associated loci. Additionally, single-cell and spatial transcriptomics have the potential to serve as cell-level readout platforms integrating cellular perturbations, enabling high-throughput discovery of causal relationships between genotype and gene expression at the cellular level in plants. Successful implementation will accelerate the identification of key genetic variants for crop improvement. Here we review lessons learned from application of single-cell screening in mammalian cells, highlight major technical and biological barriers to its use in plants, and outline potential strategies to overcome these challenges. Together, the widespread application and integration of single-cell and spatial transcriptomics with other technologies enable not only the descriptive cataloging of cell states but also the causal interrogation of sequence functions and regulatory networks at cell type resolution, ultimately advancing gene function studies and accelerating crop improvement. 

Chromatin is a hierarchically organized soft material whose nanoscale structure and heterogeneity regulate essential genomic functions. Resolving this organization requires molecular imaging materials that combine selective DNA binding with photophysical properties compatible with nanoscale localization and energy transfer under biologically relevant conditions. Conventional bisbenzimide (Hoechst) DNA stains provide robust targeting of nuclear DNA but limited applicability for super-resolution imaging. Here, we investigate the modular molecular engineering strategy in which cyanine chromophores (Cyanine3, Cyanine5, or Cyanine7) are covalently integrated with a bisbenzimide DNA-binding motif through an aliphatic spacer to form hybrid fluorescent materials. This design decouples DNA recognition from optical functionality, allowing independent optimization of binding affinity and photophysical performance. In the resulting conjugates, the cyanine units retain their intrinsic brightness and spectral properties, while the bisbenzimide ligand preserves high-affinity minor-groove binding to nuclear DNA. The Cyanine3–Hoechst and Cyanine5–Hoechst hybrids enable high-contrast imaging of nuclear DNA in fixed and permeabilized cells using long-wavelength excitation (>500 nm), with strongly suppressed extranuclear background. The substantial spectral overlap between Cyanine3 emission and Cyanine5 absorption further enables efficient Förster resonance energy transfer within the nuclear environment, providing a route to probe nanoscale proximity and organization in chromatin. Owing to their favorable photophysical stability and brightness, these hybrid materials also support single-molecule localization microscopy, revealing nanostructured features within the nucleus that are not resolved in diffraction-limited images. This approach provides a versatile platform for developing next-generation imaging materials tailored to the study of chromatin as a dynamic soft matter system. 

A<sc>bstract</sc> We construct Schubert line defects in the 3d$$ \mathcal{N}=2 $$ $N=2$ supersymmetric gauged linear sigma model (GLSM) with target space a partial flag manifoldX= Fl(k;n), generalizing our construction for complete flag manifolds given in a companion paper (part I) [1]. In the context of the 3d GLSM/quantum K-theory correspondence, the Schubert line defects are constructed as 1d$$ \mathcal{N}=2 $$ $N=2$ supersymmetric gauge theories coupled to the 3d field theory, and they flow to objects supported on Schubert varietiesX_w⊆Xin the quantum K-theory. The flavored Witten index of the 1d defect is expected to compute the Chern character of [𝒪_w] — more precisely, it gives us a polynomial representative of the Schubert class in the quantum K-theory ring. We give strong evidence for this claim by showing in examples that the Witten indices of Schubert defects indeed reproduce a recently-defined set of polynomials that represent the Schubert classes in the Whitney presentation, which we call the parabolic Whitney polynomials. Moreover, upon using the quantum ring relations, we can convert these polynomials into seemingly new polynomials in the Toda presentation, which we call the parabolic quantum Grothendieck polynomials. These new polynomials specialize to known polynomials in various limits, including to the quantum Grothendieck polynomials in the case of the complete flag. In the 2d limit, our construction also realizes the Schubert classes [X_w] in the quantum cohomology ring of the partial flag manifold, and the parabolic quantum Grothendieck polynomials then reduce to previously known parabolic quantum Schubert polynomials. 

A<sc>bstract</sc> We construct new half-BPS line defects in 3d$$ \mathcal{N}=2 $$ $N=2$ supersymmetric quiver gauge theories whose Higgs branches are complete flag manifoldsX= Fl(n). Upon circle compactification, the bulk theory flows to a non-linear sigma model (NLSM) with target spaceXand the line defects flow to objects supported on Schubert varietiesX_w⊆X. TheseSchubert line defectsform an important basis of the quantum K-theory ofX. They are realized as$$ \mathcal{N}=2 $$ $N=2$ supersymmetric quantum mechanics (SQM) quivers coupled to the 3d gauge theory. We show that the insertion of the Schubert line defect restricts the target space of the 3d gauged linear sigma model (GLSM) to the Schubert varietyX_w, with the 1d degrees of freedom physically realizing a Bott-Samelson resolution ofX_w. Moreover, we verify in examples that the 1d flavored Witten index of the quiver SQM reproduces the (equivariant) Chern character of the structure sheaf$$ {\mathcal{O}}_{X_w} $$ $O_{X_w}$ as a (double) quantum Grothendieck polynomial, generalizing previous results forXa Grassmannian manifold. Our construction thus provides a more direct realization of the 3d GLSM/quantum K-theory correspondence for complete flag manifolds. Finally, in the small-circle limit, we obtain a 0d-2d coupled system that realizes the Schubert classes [X_w] in the quantum cohomology ring ofX.