Searcharxiv⌕ Search

arXiv subjects

Chenlin Zhou

Publications and source records attributed to Chenlin Zhou.

At least 19 recordsLinked to original sources
arXiv2026

Adaptive Spiking Neurons for Vision and Language Modeling

Regarded as the third generation of neural networks, Spiking Neural Networks (SNNs) have garnered significant traction due to their biological plausibility and energy efficiency. Recent advancements in large models necessitate spiking neurons capable of high performance, adaptability, and training efficiency. In this work, we first propose a novel functional perspective that provides general guidance for designing the new generation of spiking neurons. Following the insightful guidelines, we propose the Adaptive Spiking Neuron (ASN), which incorporates trainable parameters to learn membrane potential dynamics and enable adaptive firing. ASN adopts an integer training and spike inference paradigm, facilitating efficient SNN training. To further enhance robustness, we propose a specialized variant of ASN, the Normalized Adaptive Spiking Neuron (NASN), which integrates normalization to stabilize training. We evaluate our neuron model on 19 datasets spanning five distinct tasks in both vision and language modalities, demonstrating the effectiveness and versatility of the ASN family. Our ASN family is expected to become the new generation of general-purpose spiking neurons.

cs.NE↗
arXiv2026

BiSpikCLM: A Spiking Language Model integrating Softmax-Free Spiking Attention and Spike-Aware Alignment Distillation

Spiking Neural Networks (SNNs) offer promising energy-efficient alternatives to large language models (LLMs) due to their event-driven nature and ultra-low power consumption. However, to preserve capacity, most existing spiking LLMs still incur intensive floating-point matrix multiplication (MatMul) and nonlinearities, or training difficulties arising from the complex spatiotemporal dynamics. To address these challenges, we propose BiSpikCLM, the first fully binary spiking MatMul-free causal language model. BiSpikCLM introduces Softmax-Free Spiking Attention (SFSA), eliminating softmax and floating-point operations in autoregressive language modeling. For efficient training, we introduce Spike-Aware Alignment Distillation (SpAD), which aligns ANN teacher and SNN student across embeddings, attention maps, intermediate features, and output logits. SpAD framework allows BiSpikCLM to reach comparable performance to ANN counterparts using substantially fewer training tokens (e.g., only 5.6% of the tokens for the 1.3B model). As a result, BiSpikCLM achieves competitive performance at only 4.16% - 5.87% of the computational cost on natural language generation tasks. Our results highlight the feasibility and effectiveness of fully binary spike-driven LLMs and establish the distillation as a promising pathway for brain-inspired spiking NLP.

cs.NE↗
arXiv2026

Winner-Take-All Spiking Transformer for Language Modeling

Spiking Transformers, which combine the scalability of Transformers with the sparse, energy-efficient property of Spiking Neural Networks (SNNs), have achieved impressive results in neuromorphic and vision tasks and attracted increasing attention. However, existing directly trained spiking transformers primarily focus on vision tasks. For language modeling with spiking transformer, convergence relies heavily on softmax-based spiking self-attention, which incurs high energy costs and poses challenges for neuromorphic deployment. To address this issue, we introduce Winner-Take-All (WTA) mechanisms into spiking transformers and propose two novel softmax-free, spike-driven self-attention modules: WTA Spiking Self-Attention (WSSA) and Causal WTA Spiking Self-Attention (CWSSA). Based on them, we design WTA-based Encoder-only Spiking Transformer (WE-Spikingformer) for masked language modeling and WTA-based Decoder-only Spiking Transformer (WD-Spikingformer) for causal language modeling, systematically exploring softmax-free, spiking-driven Transformer architectures trained end-to-end for natural language processing tasks. Extensive experiments on 16 datasets spanning natural language understanding, question-answering tasks, and commonsense reasoning tasks validate the effectiveness of our approach and highlight the promise of spiking transformers for general language modeling and energy-efficient artificial intelligence.

cs.NE↗
arXiv2025

SpikCommander: A High-performance Spiking Transformer with Multi-view Learning for Efficient Speech Command Recognition

Spiking neural networks (SNNs) offer a promising path toward energy-efficient speech command recognition (SCR) by leveraging their event-driven processing paradigm. However, existing SNN-based SCR methods often struggle to capture rich temporal dependencies and contextual information from speech due to limited temporal modeling and binary spike-based representations. To address these challenges, we first introduce the multi-view spiking temporal-aware self-attention (MSTASA) module, which combines effective spiking temporal-aware attention with a multi-view learning framework to model complementary temporal dependencies in speech commands. Building on MSTASA, we further propose SpikCommander, a fully spike-driven transformer architecture that integrates MSTASA with a spiking contextual refinement channel MLP (SCR-MLP) to jointly enhance temporal context modeling and channel-wise feature integration. We evaluate our method on three benchmark datasets: the Spiking Heidelberg Dataset (SHD), the Spiking Speech Commands (SSC), and the Google Speech Commands V2 (GSC). Extensive experiments demonstrate that SpikCommander consistently outperforms state-of-the-art (SOTA) SNN approaches with fewer parameters under comparable time steps, highlighting its effectiveness and efficiency for robust speech command recognition.

cs.SD↗
arXiv2025

S$^2$M-Former: Spiking Symmetric Mixing Branchformer for Brain Auditory Attention Detection

Auditory attention detection (AAD) aims to decode listeners' focus in complex auditory environments from electroencephalography (EEG) recordings, which is crucial for developing neuro-steered hearing devices. Despite recent advancements, EEG-based AAD remains hindered by the absence of synergistic frameworks that can fully leverage complementary EEG features under energy-efficiency constraints. We propose S$^2$M-Former, a novel spiking symmetric mixing framework to address this limitation through two key innovations: i) Presenting a spike-driven symmetric architecture composed of parallel spatial and frequency branches with mirrored modular design, leveraging biologically plausible token-channel mixers to enhance complementary learning across branches; ii) Introducing lightweight 1D token sequences to replace conventional 3D operations, reducing parameters by 14.7$\times$. The brain-inspired spiking architecture further reduces power consumption, achieving a 5.8$\times$ energy reduction compared to recent ANN methods, while also surpassing existing SNN baselines in terms of parameter efficiency and performance. Comprehensive experiments on three AAD benchmarks (KUL, DTU and AV-GC-AAD) across three settings (within-trial, cross-trial and cross-subject) demonstrate that S$^2$M-Former achieves comparable state-of-the-art (SOTA) decoding accuracy, making it a promising low-power, high-performance solution for AAD tasks. Code is available at https://github.com/JackieWang9811/S2M-Former.

cs.LG↗
arXiv2025

A Survey of Context Engineering for Large Language Models

The performance of Large Language Models (LLMs) is fundamentally determined by the contextual information provided during inference. This survey introduces Context Engineering, a formal discipline that transcends simple prompt design to encompass the systematic optimization of information payloads for LLMs. We present a comprehensive taxonomy decomposing Context Engineering into its foundational components and the sophisticated implementations that integrate them into intelligent systems. We first examine the foundational components: context retrieval and generation, context processing and context management. We then explore how these components are architecturally integrated to create sophisticated system implementations: retrieval-augmented generation (RAG), memory systems and tool-integrated reasoning, and multi-agent systems. Through this systematic analysis of over 1400 research papers, our survey not only establishes a technical roadmap for the field but also reveals a critical research gap: a fundamental asymmetry exists between model capabilities. While current models, augmented by advanced context engineering, demonstrate remarkable proficiency in understanding complex contexts, they exhibit pronounced limitations in generating equally sophisticated, long-form outputs. Addressing this gap is a defining priority for future research. Ultimately, this survey provides a unified framework for both researchers and engineers advancing context-aware AI.

cs.CL↗
arXiv2025

MSVIT: Improving Spiking Vision Transformer Using Multi-scale Attention Fusion

The combination of Spiking Neural Networks (SNNs) with Vision Transformer architectures has garnered significant attention due to their potential for energy-efficient and high-performance computing paradigms. However, a substantial performance gap still exists between SNN-based and ANN-based transformer architectures. While existing methods propose spiking self-attention mechanisms that are successfully combined with SNNs, the overall architectures proposed by these methods suffer from a bottleneck in effectively extracting features from different image scales. In this paper, we address this issue and propose MSVIT. This novel spike-driven Transformer architecture firstly uses multi-scale spiking attention (MSSA) to enhance the capabilities of spiking attention blocks. We validate our approach across various main datasets. The experimental results show that MSVIT outperforms existing SNN-based models, positioning itself as a state-of-the-art solution among SNN-transformer architectures. The codes are available at https://github.com/Nanhu-AI-Lab/MSViT.

cs.CV↗
arXiv2024

Efficient Speech Command Recognition Leveraging Spiking Neural Network and Curriculum Learning-based Knowledge Distillation

The intrinsic dynamics and event-driven nature of spiking neural networks (SNNs) make them excel in processing temporal information by naturally utilizing embedded time sequences as time steps. Recent studies adopting this approach have demonstrated SNNs' effectiveness in speech command recognition, achieving high performance by employing large time steps for long time sequences. However, the large time steps lead to increased deployment burdens for edge computing applications. Thus, it is important to balance high performance and low energy consumption when detecting temporal patterns in edge devices. Our solution comprises two key components. 1). We propose a high-performance fully spike-driven framework termed SpikeSCR, characterized by a global-local hybrid structure for efficient representation learning, which exhibits long-term learning capabilities with extended time steps. 2). To further fully embrace low energy consumption, we propose an effective knowledge distillation method based on curriculum learning (KDCL), where valuable representations learned from the easy curriculum are progressively transferred to the hard curriculum with minor loss, striking a trade-off between power efficiency and high performance. We evaluate our method on three benchmark datasets: the Spiking Heidelberg Dataset (SHD), the Spiking Speech Commands (SSC), and the Google Speech Commands (GSC) V2. Our experimental results demonstrate that SpikeSCR outperforms current state-of-the-art (SOTA) methods across these three datasets with the same time steps. Furthermore, by executing KDCL, we reduce the number of time steps by 60% and decrease energy consumption by 54.8% while maintaining comparable performance to recent SOTA results. Therefore, this work offers valuable insights for tackling temporal processing challenges with long time sequences in edge neuromorphic computing systems.

cs.LG↗
arXiv2024

Direct Training High-Performance Deep Spiking Neural Networks: A Review of Theories and Methods

Spiking neural networks (SNNs) offer a promising energy-efficient alternative to artificial neural networks (ANNs), in virtue of their high biological plausibility, rich spatial-temporal dynamics, and event-driven computation. The direct training algorithms based on the surrogate gradient method provide sufficient flexibility to design novel SNN architectures and explore the spatial-temporal dynamics of SNNs. According to previous studies, the performance of models is highly dependent on their sizes. Recently, direct training deep SNNs have achieved great progress on both neuromorphic datasets and large-scale static datasets. Notably, transformer-based SNNs show comparable performance with their ANN counterparts. In this paper, we provide a new perspective to summarize the theories and methods for training deep SNNs with high performance in a systematic and comprehensive way, including theory fundamentals, spiking neuron models, advanced SNN models and residual architectures, software frameworks and neuromorphic hardware, applications, and future trends. The reviewed papers are collected at https://github.com/zhouchenlin2096/Awesome-Spiking-Neural-Networks

cs.NE↗
arXiv2024

SVFormer: A Direct Training Spiking Transformer for Efficient Video Action Recognition

Video action recognition (VAR) plays crucial roles in various domains such as surveillance, healthcare, and industrial automation, making it highly significant for the society. Consequently, it has long been a research spot in the computer vision field. As artificial neural networks (ANNs) are flourishing, convolution neural networks (CNNs), including 2D-CNNs and 3D-CNNs, as well as variants of the vision transformer (ViT), have shown impressive performance on VAR. However, they usually demand huge computational cost due to the large data volume and heavy information redundancy introduced by the temporal dimension. To address this challenge, some researchers have turned to brain-inspired spiking neural networks (SNNs), such as recurrent SNNs and ANN-converted SNNs, leveraging their inherent temporal dynamics and energy efficiency. Yet, current SNNs for VAR also encounter limitations, such as nontrivial input preprocessing, intricate network construction/training, and the need for repetitive processing of the same video clip, hindering their practical deployment. In this study, we innovatively propose the directly trained SVFormer (Spiking Video transFormer) for VAR. SVFormer integrates local feature extraction, global self-attention, and the intrinsic dynamics, sparsity, and spike-driven nature of SNNs, to efficiently and effectively extract spatio-temporal features. We evaluate SVFormer on two RGB datasets (UCF101, NTU-RGBD60) and one neuromorphic dataset (DVS128-Gesture), demonstrating comparable performance to the mainstream models in a more efficient way. Notably, SVFormer achieves a top-1 accuracy of 84.03% with ultra-low power consumption (21 mJ/video) on UCF101, which is state-of-the-art among directly trained deep SNNs, showcasing significant advantages over prior models.

cs.CV↗
arXiv2024

QKFormer: Hierarchical Spiking Transformer using Q-K Attention

Spiking Transformers, which integrate Spiking Neural Networks (SNNs) with Transformer architectures, have attracted significant attention due to their potential for energy efficiency and high performance. However, existing models in this domain still suffer from suboptimal performance. We introduce several innovations to improve the performance: i) We propose a novel spike-form Q-K attention mechanism, tailored for SNNs, which efficiently models the importance of token or channel dimensions through binary vectors with linear complexity. ii) We incorporate the hierarchical structure, which significantly benefits the performance of both the brain and artificial neural networks, into spiking transformers to obtain multi-scale spiking representation. iii) We design a versatile and powerful patch embedding module with a deformed shortcut specifically for spiking transformers. Together, we develop QKFormer, a hierarchical spiking transformer based on Q-K attention with direct training. QKFormer shows significantly superior performance over existing state-of-the-art SNN models on various mainstream datasets. Notably, with comparable size to Spikformer (66.34 M, 74.81%), QKFormer (64.96 M) achieves a groundbreaking top-1 accuracy of 85.65% on ImageNet-1k, substantially outperforming Spikformer by 10.84%. To our best knowledge, this is the first time that directly training SNNs have exceeded 85% accuracy on ImageNet-1K. The code and models are publicly available at https://github.com/zhouchenlin2096/QKFormer

cs.NE↗
arXiv2023

Enhancing the Performance of Transformer-based Spiking Neural Networks by SNN-optimized Downsampling with Precise Gradient Backpropagation

Deep spiking neural networks (SNNs) have drawn much attention in recent years because of their low power consumption, biological rationality and event-driven property. However, state-of-the-art deep SNNs (including Spikformer and Spikingformer) suffer from a critical challenge related to the imprecise gradient backpropagation. This problem arises from the improper design of downsampling modules in these networks, and greatly hampering the overall model performance. In this paper, we propose ConvBN-MaxPooling-LIF (CML), an SNN-optimized downsampling with precise gradient backpropagation. We prove that CML can effectively overcome the imprecision of gradient backpropagation from a theoretical perspective. In addition, we evaluate CML on ImageNet, CIFAR10, CIFAR100, CIFAR10-DVS, DVS128-Gesture datasets, and show state-of-the-art performance on all these datasets with significantly enhanced performances compared with Spikingformer. For instance, our model achieves 77.64 $\%$ on ImageNet, 96.04 $\%$ on CIFAR10, 81.4$\%$ on CIFAR10-DVS, with + 1.79$\%$ on ImageNet, +1.16$\%$ on CIFAR100 compared with Spikingformer.

cs.NE↗
arXiv2023

Spikingformer: A Key Foundation Model for Spiking Neural Networks

Spiking neural networks (SNNs) offer a promising energy-efficient alternative to artificial neural networks, due to their event-driven spiking computation. However, some foundation SNN backbones (including Spikformer and SEW ResNet) suffer from non-spike computations (integer-float multiplications) caused by the structure of their residual connections. These non-spike computations increase SNNs' power consumption and make them unsuitable for deployment on mainstream neuromorphic hardware. In this paper, we analyze the spike-driven behavior of the residual connection methods in SNNs. We then present Spikingformer, a novel spiking transformer backbone that merges the MS Residual connection with Self-Attention in a biologically plausible way to address the non-spike computation challenge in Spikformer while maintaining global modeling capabilities. We evaluate Spikingformer across 13 datasets spanning large static images, neuromorphic data, and natural language tasks, and demonstrate the effectiveness and universality of Spikingformer, setting a vital benchmark for spiking neural networks. In addition, with the spike-driven features and global modeling capabilities, Spikingformer is expected to become a more efficient general-purpose SNN backbone towards energy-efficient artificial intelligence. Code: https://github.com/TheBrainLab/Spikingformer

cs.NE↗
arXiv2023

Learning Tri-mode Grasping for Ambidextrous Robot Picking

Object picking in cluttered scenes is a widely investigated field of robot manipulation, however, ambidextrous robot picking is still an important and challenging issue. We found the fusion of different prehensile actions (grasp and suction) can expand the range of objects that can be picked by robot, and the fusion of prehensile action and nonprehensile action (push) can expand the picking space of ambidextrous robot. In this paper, we propose a Push-Grasp-Suction (PGS) tri-mode grasping learning network for ambidextrous robot picking through the fusion of different prehensile actions and the fusion of prehensile action and nonprehensile aciton. The prehensile branch of PGS takes point clouds as input, and the 6-DoF picking configuration of grasp and suction in cluttered scenes are generated by multi-task point cloud learning. The nonprehensile branch with depth image input generates instance segmentation map and push configuration, cooperating with the prehensile actions to complete the picking of objects out of single-arm space. PGS generalizes well in real scene and achieves state-of-the-art picking performance.

cs.RO↗
arXiv2022

Chemical exploration of Galactic cold cores

A solar-type system starts from an initial molecular core that acquires organic complexity as it evolves. The so-called prestellar cores that can be studied are rare, which has hampered our understanding of how organic chemistry sets in and grows. Aims. We selected the best prestellar core targets from the cold core catalogue that represent a diversity in terms of their environment to explore their chemical complexity: 1390 (in the compressed shell of Lambda Ori), 869 (in the MBM12 cloud), and 4149 (in the California nebula). We obtained a spectral survey with the IRAM 30 m telescope in order to explore the molecular complexity of the cores. We carried out a radiative transfer analysis of the detected transitions in order to place some constraints on the physical conditions of the cores and on the molecular column densities. We also used the molecular ions in the survey to estimate the cosmic-ray ionisation rate and the S/H initial elemental abundance using a gas-phase chemical model to reproduce their abundances. We found large differences in the molecular complexity (deuteration, complex organic molecules, sulphur, carbon chains, and ions) and compared their chemical properties with a cold core and two prestellar cores. The chemical diversity we found in the three cores seems to be correlated with their chemical evolution: two of them are prestellar (1390 and 4149), and one is in an earlier stage (869). The influence of the environment is likely limited because cold cores are strongly shielded from their surroundings. The high extinction prevents interstellar UV radiation from penetrating deeply into the cores. Higher spatial resolution observations of the cores are therefore needed to constrain the physical structure of the cores, as well as a larger-scale distribution of molecular ions to understand the influence of the environment on their molecular complexity.

astro-ph.GA↗
arXiv2020

Planck Galactic Cold Clumps in Two Regions: the First Quadrant and the Anti-Center Direction Region

Sixty five Planck Galactic cold clumps (PGCCs) from the first quadrant (IQuad) and thirty nine of PGCCs from the Anti-Center direction region (ACent) were observed in $^{12}$CO, $^{13}$CO and C$^{18}$O J=1-0 lines using the PMO 13.7-m telescope. All the targets were detected with all the three lines, except for 12 IQuad and 8 ACent PGCCs without C$^{18}$O detection. Seventy six and 49 velocity components were obtained in IQuad and ACent respectively. One-hundred and forty-six cores were extracted from 76 IQuad clumps and 100 cores from 49 ACent clumps. The average T$_{\mathrm{ex}}$ of IQuad cores and ACent cores are 12.4 K and 12.1 K, respectively. The average line width of $^{13}$CO of IQuad cores and ACent cores are 1.55 km s$^{-1}$ and 1.77 km s$^{-1}$, respectively. Among the detected cores, 24 in IQuad and 13 in ACent have asymmetric line profiles. The small blue excesses, $\sim$0.03 in IQuad and 0.01 in ACent, indicate that the star formation is not active in these PGCC cores. Power-law fittings of core mass function to the high mass end give indexes of -0.57 in IQuad and -1.02 in ACent which are flatter than the slope of initial mass function given by \citeauthor{1955ApJ...121..161S}. The large turnover masses with value of 28 M$_{\odot}$ for IQuad cores and 77 M$_{\odot}$ for ACent cores suggest low star formation efficiencies in PGCCs. The correlation between virial mass and gas mass indicates that most of PGCC cores in both regions are not likely pressure-confined.

astro-ph.GA↗
arXiv2019

Using CO line ratios to trace compressed areas in bubble N131

N131 is a typical infrared dust bubble showing an expanding ringlike shell. We study what kinds of CO line ratios can be used to trace the interaction in the expanding bubble. We carry out new $\rm CO\,(3-2)$ observations towards bubble N131 using the JCMT 15-m telescope, and derive line ratios by combining with our previous $\rm CO\,(2-1)$ and $\rm CO\,(1-0)$ data from the IRAM 30-m telescope observations. To trace the interaction between the molecular gas and the ionized gas in the HII region, we use RADEX to model the dependence of CO line ratios on kinetic temperature and H$_2$ volume density, and examine the abnormal line ratios based on other simulations. We present $\rm CO\,(3-2)$, $\rm CO\,(2-1)$, and $\rm CO\,(1-0)$ integrated intensity maps convolved to the same angular resolution (22.5$"$). The three different CO transition maps show apparently similar morphology. The line ratios of $W_{\rm CO\,(3-2)}$/$W_{\rm CO\,(2-1)}$ mostly range from 0.2 to 1.2 with a median of $0.54\pm0.12$, while the line ratios of $W_{\rm CO\,(2-1)}$/$W_{\rm CO\,(1-0)}$ range from 0.5 to 1.6 with a median of $0.84\pm0.15$. The high CO line ratios $W_{\rm CO\,(3-2)}$/$W_{\rm CO\,(2-1)}\gtrsim 0.8 $ and $W_{\rm CO\,(2-1)}$/$W_{\rm CO\,(1-0)}\gtrsim 1.2$ are beyond the threshold predicted by numerical simulations based on the assumed density-temperature structure for the inner rims of ringlike shell, where are the compressed areas in bubble N131. These high CO integrated intensity ratios, such as $W_{\rm CO\,(3-2)}$/$W_{\rm CO\,(2-1)}\gtrsim0.8$ and $W_{\rm CO\,(2-1)}$/$W_{\rm CO\,(1-0)}\gtrsim1.2$, can be used as a tracer of gas compressed regions with a relatively high temperature and density. This further suggests that the non-Gaussian part of the line-ratio distribution can be used to trace the interaction between the molecular gas and the hot gas in the bubble.

astro-ph.GA↗