Search for: All records

We present iSeg, a new interactive technique for segmenting 3D shapes. Previous works have focused mainly on leveraging pre-trained 2D foundation models for 3D segmentation based on text. However, text may be insufficient for accurately describing fine-grained spatial segmentations. Moreover, achieving a consistent 3D segmentation using a 2D model is highly challenging, since occluded areas of the same semantic region may not be visible together from any 2D view. Thus, we design a segmentation method conditioned on fine user clicks, which operates entirely in 3D. Our system accepts user clicks directly on the shape's surface, indicating the inclusion or exclusion of regions from the desired shape partition. To accommodate various click settings, we propose a novel interactive attention module capable of processing different numbers and types of clicks, enabling the training of a single unified interactive segmentation model. We apply iSeg to a myriad of shapes from different domains, demonstrating its versatility and faithfulness to the user's specifications. Our project page is at https://threedle.github.io/iSeg/. 

Abstract The light chain of tetanus neurotoxin (TeNT) is a 52 kD metalloprotease that potently inhibits synaptic transmission by cleaving the endogenous vesicle fusion protein VAMP2. To mitigate the toxicity of TeNT and harness it as a conditional tool for neuroscience, we engineered Light-Activated TeNT (LATeNT) via insertion of the light-sensitive LOV domain into an allosteric site. LATeNT was optimized by directed evolution and shown to have undetectable activity in the dark mammalian brain. Following 30 seconds of weak blue light exposure, however, LATeNT potently inhibited synaptic transmission in multiple brain regions. The effect could be reversed over 24 hours. We used LATeNT to discover an interneuron population in hippocampus that controls anxiety-like behaviors in mouse, and to control the secretion of endogenous insulin from pancreatic beta cells. Synthetic circuits incorporating LATeNT converted drug, Ca²⁺, or receptor activation into transgene expression or reporter protein secretion. Due to its large dynamic range, rapid kinetics, and highly specific mechanism of action, LATeNT should be a robust tool for conditional proteolysis and spatiotemporal control of synaptic transmissionin vivo. 

With the wide adoption of deep neural network (DNN) models for various applications, enterprises, and cloud providers have built deep learning clusters and increasingly deployed specialized accelerators, such as GPUs and TPUs, for DNN training jobs. To arbitrate cluster resources among multi-user jobs, existing schedulers fall short, either lacking fine-grained heterogeneity awareness or hardly generalizable to various scheduling policies. To fill this gap, we propose a novel design of a task-level heterogeneity-aware scheduler, Hadar, based on an online optimization framework that can express other scheduling algorithms. Hadar leverages the performance traits of DNN jobs on a heterogeneous cluster, characterizes the task-level performance heterogeneity in the optimization problem, and makes scheduling decisions across both spatial and temporal dimensions. The primal-dual framework is employed, with our design of a dual subroutine, to solve the optimization problem and guide the scheduling design. Extensive trace-driven simulations with representative DNN models have been conducted to demonstrate that Hadar improves the average job completion time (JCT) by 3× over an Apache YARN-based resource manager used in production. Moreover, Hadar outperforms Gavel[1], the state-of-the-art heterogeneity-aware scheduler, by 2.5× for the average JCT, and shortens the queuing delay by 13% and improve FTF (Finish-Time-Fairness) by 1.5%.