Search for: All records

Summary In both plants and animals, tissue or organ regeneration typically follows wounding, which also activates defense responses against pathogenic microbes and herbivores. Both intrinsic and environmental cues guide the molecular decisions between regeneration and defense. In animal studies, extensive research has highlighted the role of various microbes – including pathogenic, commensal, and beneficial species – in influencing the signaling interplay between immunity and regeneration. Conversely, most plant regeneration studies are conducted under sterile conditions, which leaves a gap in our understanding of how plant innate immunity influences regeneration pathways. Recent findings have begun to elucidate the roles of key defense pathways in modulating plant regeneration and the crosstalk between these two processes. These studies also explore how microbes might influence the molecular choice between defense and regeneration in plants. This review examines the molecular mechanisms governing the balance between plant regeneration and innate immunity, with a focus on the emerging role of aging and microbial interactions in shaping these processes. 

The non-volatile Resistive RAM (ReRAM) crossbar has shown great potential in accelerating inference in various machine learning models However, it suffers from high reprogramming energy, hindering its usage for on-device adaption to new tasks. Recently, parameter-efficient fine-tuning methods, such as Low-Rank Adaption (LoRA), have been proposed to train few parameters while matching full fine-tuning performance. However, in ReRAM crossbar, the reprogramming cost of LoRA is non-trivial and will increase significantly when adapting to multi-tasks on the device. To address this issue, we are the first to propose LoRAFusion, a parameter-efficient multi-task on-device learning framework for ReRAM crossbar via fusion of pre-trained LoRA modules. LoRAFusion is a group of LoRA modules that are one-time learned based on diverse domain-specific tasks and deployed to the crossbar, acting as the pool of background knowledge. Then given a new unseen task, those LoRA modules are frozen (i.e., no energy-hungry ReRAM cells reprograming), only the proposed learnable layer-wise LoRA fusion coefficient and magnitude vector parameters are trained on-device to weighted-combine pre-trained LoRA modules, which significantly reduces the training parameter size. Our comprehensive experiments show LoRAFusion only uses 3% of the number of trainable parameters in LoRA (148K vs. 4700K), with 0.19% accuracy drop. Codes are available at https://github.com/ASU-ESIC-FAN-Lab/LoRAFusion 

Abstract The recent discovery of topological flat bands in twisted transition metal dichalcogenide homobilayers and multilayer graphene has sparked significant research interest. Here, a new platform for realizing tunable topological moiré flat bands: twisted type‐II Rashba homobilayers, is proposed. By maintaining centrosymmetry, the interplay between Rashba spin‐orbit coupling and interlayer interactions generates an effective pseudo‐antiferromagnetic field, opening a gap within the Dirac cone with non‐zero Berry curvature. Using twisted BiTeI bilayers as an example, it is predicted that the emergence of flat topological bands with a remarkably narrow bandwidth (below 20 meV). Notably, the system undergoes a transition from a valley Hall insulator to a quantum spin Hall insulator as the twisting angle increases. This transition arises from a competition between the twisting‐driven effective spin‐orbit coupling and sublattice onsite energies presented in type‐II Rashba moiré structures. The high tunability of Rashba materials in terms of the spin‐orbit coupling strength, interlayer interaction, and twisting angle expands the range of materials suitable for functionalizing and manipulating correlated topological properties.