Qiang Yu

NeurIPS Conference 2025 Conference Paper

HetSyn: Versatile Timescale Integration in Spiking Neural Networks via Heterogeneous Synapses

• Zhichao Deng
• Zhikun Liu
• Junxue Wang
• Shengqian Chen
• Xiang Wei
• Qiang Yu

Spiking Neural Networks (SNNs) offer a biologically plausible and energy-efficient framework for temporal information processing. However, existing studies overlook a fundamental property widely observed in biological neurons—synaptic heterogeneity, which plays a crucial role in temporal processing and cognitive capabilities. To bridge this gap, we introduce HetSyn, a generalized framework that models synaptic heterogeneity with synapse-specific time constants. This design shifts temporal integration from the membrane potential to the synaptic current, enabling versatile timescale integration and allowing the model to capture diverse synaptic dynamics. We implement HetSyn as HetSynLIF, an extended form of the leaky integrate-and-fire (LIF) model equipped with synapse-specific decay dynamics. By adjusting the parameter configuration, HetSynLIF can be specialized into vanilla LIF neurons, neurons with threshold adaptation, and neuron-level heterogeneous models. We demonstrate that HetSynLIF not only improves the performance of SNNs across a variety of tasks—including pattern generation, delayed match-to-sample, speech recognition, and visual recognition—but also exhibits strong robustness to noise, enhanced working memory performance, efficiency under limited neuron resources, and generalization across timescales. In addition, analysis of the learned synaptic time constants reveals trends consistent with empirical observations in biological synapses. These findings underscore the significance of synaptic heterogeneity in enabling efficient neural computation, offering new insights into brain-inspired temporal modeling. Code available at: https: //github. com/dzcgood/HetSyn.

NeurIPS Conference 2025 Conference Paper

Seemingly Redundant Modules Enhance Robust Odor Learning in Fruit Flies

• HaiYang Li
• Liao Yu
• Qiang Yu
• Yunliang Zang

Biological circuits have evolved to incorporate multiple modules that perform similar functions. In the fly olfactory circuit, both lateral inhibition (LI) and neuronal spike frequency adaptation (SFA) are thought to enhance pattern separation for odor learning. However, it remains unclear whether these mechanisms play redundant or distinct roles in this process. In this study, we present a computational model of the fly olfactory circuit to investigate odor discrimination under varying noise conditions that simulate complex environments. Our results show that LI primarily enhances odor discrimination in low‑ and medium‑noise scenarios, but this benefit diminishes and may reverse under higher‑noise conditions. In contrast, SFA consistently improves discrimination across all noise levels. LI is preferentially engaged in low‑ and medium‑noise environments, whereas SFA dominates in high‑noise settings. When combined, these two sparsification mechanisms enable optimal discrimination performance. This work demonstrates that seemingly redundant modules in biological circuits can, in fact, be essential for achieving optimal learning in complex contexts.

EAAI Journal 2024 Journal Article

Deep convolutional neural network based on self-distillation for tool wear recognition

• Yi Pan
• Ling Hao
• Jianliang He
• Kun Ding
• Qiang Yu
• Yulin Wang

AAAI Conference 2024 Conference Paper

Weak Distribution Detectors Lead to Stronger Generalizability of Vision-Language Prompt Tuning

• Kun Ding
• Haojian Zhang
• Qiang Yu
• Ying Wang
• Shiming Xiang
• Chunhong Pan

We propose a generalized method for boosting the generalization ability of pre-trained vision-language models (VLMs) while fine-tuning on downstream few-shot tasks. The idea is realized by exploiting out-of-distribution (OOD) detection to predict whether a sample belongs to a base distribution or a novel distribution and then using the score generated by a dedicated competition based scoring function to fuse the zero-shot and few-shot classifier. The fused classifier is dynamic, which will bias towards the zero-shot classifier if a sample is more likely from the distribution pre-trained on, leading to improved base-to-novel generalization ability. Our method is performed only in test stage, which is applicable to boost existing methods without time-consuming re-training. Extensive experiments show that even weak distribution detectors can still improve VLMs' generalization ability. Specifically, with the help of OOD detectors, the harmonic mean of CoOp and ProGrad increase by 2.6 and 1.5 percentage points over 11 recognition datasets in the base-to-novel setting.

AAAI Conference 2021 Conference Paper

Consensus Graph Representation Learning for Better Grounded Image Captioning

• Wenqiao Zhang
• Haochen Shi
• Siliang Tang
• Jun Xiao
• Qiang Yu
• Yueting Zhuang

The contemporary visual captioning models frequently hallucinate objects that are not actually in a scene, due to the visual misclassification or over-reliance on priors that resulting in the semantic inconsistency between the visual information and the target lexical words. The most common way is to encourage the captioning model to dynamically link generated object words or phrases to appropriate regions of the image, i. e. , the grounded image captioning (GIC). However, GIC utilizes an auxiliary task (grounding objects) that has not solved the key issue of object hallucination, i. e. , the semantic inconsistency. In this paper, we take a novel perspective on the issue above: exploiting the semantic coherency between the visual and language modalities. Specifically, we propose the Consensus Rraph Representation Learning framework (CGRL) for GIC that incorporates a consensus representation into the grounded captioning pipeline. The consensus is learned by aligning the visual graph (e. g. , scene graph) to the language graph that consider both the nodes and edges in a graph. With the aligned consensus, the captioning model can capture both the correct linguistic characteristics and visual relevance, and then grounding appropriate image regions further. We validate the effectiveness of our model, with a significant decline in object hallucination (-9% CHAIRi) on the Flickr30k Entities dataset. Besides, our CGRL also evaluated by several automatic metrics and human evaluation, the results indicate that the proposed approach can simultaneously improve the performance of image captioning (+2. 9 Cider) and grounding (+2. 3 F1LOC ).

AAAI Conference 2021 Conference Paper

Disentangled Motif-aware Graph Learning for Phrase Grounding

• Zongshen Mu
• Siliang Tang
• Jie Tan
• Qiang Yu
• Yueting Zhuang

In this paper, we propose a novel graph learning framework for phrase grounding in the image. Developing from the sequential to the dense graph model, existing works capture coarse-grained context but fail to distinguish the diversity of context among phrases and image regions. In contrast, we pay special attention to different motifs implied in the context of the scene graph and devise the disentangled graph network to integrate the motif-aware contextual information into representations. Besides, we adopt interventional strategies at the feature and the structure levels to consolidate and generalize representations. Finally, the cross-modal attention network is utilized to fuse intra-modal features, where each phrase can be computed similarity with regions to select the bestgrounded one. We validate the efficiency of disentangled and interventional graph network (DIGN) through a series of ablation studies, and our model achieves state-of-the-art performance on Flickr30K Entities and ReferIt Game benchmarks.

AAAI Conference 2020 Conference Paper

New Efficient Multi-Spike Learning for Fast Processing and Robust Learning

• Shenglan Li
• Qiang Yu

Spiking neural networks (SNNs) are considered to be more biologically plausible and lower power consuming than traditional artificial neural networks (ANNs). SNNs use discrete spikes as input and output, but how to process and learn these discrete spikes efficiently and accurately still remains a challenging task. Moreover, most existing learning methods are inefficient with complicated neuron dynamics and learning procedures being involved. In this paper, we propose efficient alternatives by firstly introducing a simplified and efficient neuron model. Based on it, we develop two new multi-spike learning rules together with an eventdriven scheme being presented to improve the processing ef- ficiency. We show that, with the as-proposed rules, a single neuron can be trained to successfully perform challenging tasks such as multi-category classification and feature extraction. Our learning methods demonstrate a significant robustness against various strong noises. Moreover, experimental results on some real-world classification tasks show that our approaches yield higher efficiency with less requirement on computation resource, highlighting the advantages and potential of spike-based processing and driving more efforts towards neuromorphic computing.

IJCAI Conference 2019 Conference Paper

Fast and Accurate Classification with a Multi-Spike Learning Algorithm for Spiking Neurons

• Rong Xiao
• Qiang Yu
• Rui Yan
• Huajin Tang

The formulation of efficient supervised learning algorithms for spiking neurons is complicated and remains challenging. Most existing learning methods with the precisely firing times of spikes often result in relatively low efficiency and poor robustness to noise. To address these limitations, we propose a simple and effective multi-spike learning rule to train neurons to match their output spike number with a desired one. The proposed method will quickly find a local maximum value (directly related to the embedded feature) as the relevant signal for synaptic updates based on membrane potential trace of a neuron, and constructs an error function defined as the difference between the local maximum membrane potential and the firing threshold. With the presented rule, a single neuron can be trained to learn multi-category tasks, and can successfully mitigate the impact of the input noise and discover embedded features. Experimental results show the proposed algorithm has higher precision, lower computation cost, and better noise robustness than current state-of-the-art learning methods under a wide range of learning tasks.

YNIMG Journal 2018 Journal Article

Can anomalous diffusion models in magnetic resonance imaging be used to characterise white matter tissue microstructure?

• Qiang Yu
• David Reutens
• Viktor Vegh

ICRA Conference 2014 Conference Paper

RGMP-ROS: A real-time ROS architecture of hybrid RTOS and GPOS on multi-core processor

• Hongxing Wei
• Zhen Huang
• Qiang Yu
• Miao Liu
• Yong Guan
• Jindong Tan

Recently, the open-source robot operating system (ROS) has been growing rapidly in the robotics community. However, the ROS runs on Linux, which does not provide timing guarantees for robot motion. This paper present a hybrid real-time ROS architecture on multi-core processor “RGMP-ROS”, which consists of two parts including the non-real-time subsystem “GPOS (General Operating system)” and the real-time one “RTOS (Real-time Operating system)”. The GPOS is comprised of non-real-time ROS nodes running in Linux, while the RTOS only contains real-time ROS nodes running in Nuttx. To get higher operational efficiency, the RGMP-ROS system is executed by a dual-core processor, one CPU for GPOS and the other for RTOS. The RGMP-ROS has used in the controller of a 6-DOF modular manipulator, and its effectiveness and efficiency are demonstrated by software testing and experiments. The main contributions of the present work lie in the realization of real-time ROS architecture and the application of multi-core processor in the hybrid control of an industrial robot.