EgoNeMo: Transferable Map of Pedestrian Dynamics via Egocentric LiDAR Scan
本文提出一种基于自我中心3D LiDAR点云的可迁移动态地图框架EgoNeMo，以解决传统方法需要特定地点轨迹积累的问题，通过平衡采样策略和多任务学习架构提高行人轨迹预测的准确性。
0 citationsRead paper
Selected work
Representative Papers
BiMTokenizer: Preserving Semantic-Acoustic Balance in Low-Bitrate Speech Tokenization via Bidirectional State-Space Modeling
为解决语音编码中语义保持与声学保真度之间的冲突，提出BiMTokenizer，采用双向状态空间模型和残余球形Leech量化方法，在低比特率下实现平衡。
0 citationsRead paper
Performance Evaluation of RED-ONION: A High-Speed Disk-to-Disk Transfer System
为解决科研数据传输瓶颈问题，本文介绍了一种名为RED-ONION的高速磁盘到磁盘传输系统，通过结合专用网络和优化软件实现了高效的数据传输。
0 citationsRead paper
TopoAgent: A Structure-Aware Perception-to-Reasoning Framework for Diagram-to-Graph Topology Extraction with Large Vision-Language Models
本文提出TopoAgent框架，通过结合感知和推理解决从结构图中提取实体及其连接的拓扑图的问题，使用大型视觉-语言模型在新基准TopoBench-180上实现更优性能。
0 citationsRead paper
Scalable Training of Continuous-Time Spiking Neural Networks with Differentiable Spike-Time Discretization
This work addresses the prohibitive memory overhead of deep continuous-time spiking neural networks (SNNs), which arises from the precise tracking of spike timings and hinders scalability. To overcome this limitation, the authors propose a Differentiable Spike Time Discretization (DSTD) framework that maps irregular presynaptic spikes onto fixed time steps as differentiable weighted events, accurately approximating continuous membrane potential dynamics while drastically reducing memory consumption. By integrating the leaky integrate-and-fire (LIF) neuron model, time-to-first-spike (TTFS) encoding, and a temporal regularization mechanism inspired by synchronous firing chains, the approach effectively mitigates neuronal death and enables pipeline-like training. Experiments demonstrate successful training of a 9-layer CIFAR-10 and a 20-layer Fashion-MNIST convolutional SNN on a single GPU, achieving approximately 100× lower peak memory usage and 20× faster training speed.
0 citationsRead paper
Recent publications
Latest Papers
EgoNeMo: Transferable Map of Pedestrian Dynamics via Egocentric LiDAR Scan
本文提出一种基于自我中心3D LiDAR点云的可迁移动态地图框架EgoNeMo，以解决传统方法需要特定地点轨迹积累的问题，通过平衡采样策略和多任务学习架构提高行人轨迹预测的准确性。
0 citationsRead paper
BiMTokenizer: Preserving Semantic-Acoustic Balance in Low-Bitrate Speech Tokenization via Bidirectional State-Space Modeling
为解决语音编码中语义保持与声学保真度之间的冲突，提出BiMTokenizer，采用双向状态空间模型和残余球形Leech量化方法，在低比特率下实现平衡。
0 citationsRead paper
Performance Evaluation of RED-ONION: A High-Speed Disk-to-Disk Transfer System
为解决科研数据传输瓶颈问题，本文介绍了一种名为RED-ONION的高速磁盘到磁盘传输系统，通过结合专用网络和优化软件实现了高效的数据传输。
0 citationsRead paper
TopoAgent: A Structure-Aware Perception-to-Reasoning Framework for Diagram-to-Graph Topology Extraction with Large Vision-Language Models
本文提出TopoAgent框架，通过结合感知和推理解决从结构图中提取实体及其连接的拓扑图的问题，使用大型视觉-语言模型在新基准TopoBench-180上实现更优性能。
0 citationsRead paper
Scalable Training of Continuous-Time Spiking Neural Networks with Differentiable Spike-Time Discretization
This work addresses the prohibitive memory overhead of deep continuous-time spiking neural networks (SNNs), which arises from the precise tracking of spike timings and hinders scalability. To overcome this limitation, the authors propose a Differentiable Spike Time Discretization (DSTD) framework that maps irregular presynaptic spikes onto fixed time steps as differentiable weighted events, accurately approximating continuous membrane potential dynamics while drastically reducing memory consumption. By integrating the leaky integrate-and-fire (LIF) neuron model, time-to-first-spike (TTFS) encoding, and a temporal regularization mechanism inspired by synchronous firing chains, the approach effectively mitigates neuronal death and enables pipeline-like training. Experiments demonstrate successful training of a 9-layer CIFAR-10 and a 20-layer Fashion-MNIST convolutional SNN on a single GPU, achieving approximately 100× lower peak memory usage and 20× faster training speed.
0 citationsRead paper