Research map / Time series / Long-term forecasting
Enhancing Adaptive History Reserving by Spiking Convolutional Block Attention Module in Recurrent Neural Networks
Qi Xu, Yuyuan Gao, Jiangrong Shen, Yaxin Li, Xuming Ran, Huajin Tang, Gang Pan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add Spiking CBAM to the forget gate (ablation) | Spiking ConvLSTM · DVS128-Gesture | accuracy (%): 93.3 -> 95.5 | Improved |
| Use Spiking CBAM rather than conventional CBAM on the forget gate (ablation) | Spiking ConvLSTM · DVS128-Gesture | accuracy (%): 92.4 -> 95.5 | Improved |
| Apply Spiking CBAM to all three gates rather than the forget gate alone (ablation) | Spiking ConvLSTM · DVS128-Gesture | accuracy (%): 95.5 -> 88.8 | Hurt |
| Apply Spiking CBAM to input and output gates instead of the forget gate (ablation) | Spiking ConvLSTM · DVS128-Gesture | accuracy (%): 95.5 -> 86.1 | Hurt |
| Add Spiking CBAM to the input gate as well as the forget gate (ablation) | Spiking ConvLSTM · DVS128-Gesture | accuracy (%): 95.5 -> 92.0 | Hurt |
| Apply Spiking CBAM to the input gate only (ablation) | Spiking ConvLSTM · DVS128-Gesture | accuracy (%): 93.3 -> 91.3 | Hurt |
| Use a trainable MultiStepLIFNode threshold rather than a trainable LIFNode threshold (ablation) | RSNN-SCBAM · DVS128-Gesture | accuracy (%): 91.3 -> 89.9 | Hurt |
| Set the time step to 10 instead of the reported optimal 20 (ablation) | RSNN-SCBAM · DVS128-Gesture | accuracy (%): 89.2 | Hurt |
| Set the time step to 15 instead of the reported optimal 20 (ablation) | RSNN-SCBAM · DVS128-Gesture | accuracy (%): 92.4 | Hurt |
| Set the time step to 25 instead of the reported optimal 20 (ablation) | RSNN-SCBAM · DVS128-Gesture | accuracy (%): 90.3 | Hurt |
| Use the proposed model instead of LIF-NET | CIFAR10-DVS classifier · CIFAR10-DVS | accuracy (%): 63.5 -> 66.3 | Improved |
| Use the proposed model instead of LIAF-NET | CIFAR10-DVS classifier · CIFAR10-DVS | accuracy (%): 70.4 -> 66.3 | Hurt |
Closest papers by meaning
- STAA-SNN: Spatial-Temporal Attention Aggregator for Spiking Neural Networks (CVPR 2025 · similarity 0.86)
- Learning Delays in Spiking Neural Networks using Dilated Convolutions with Learnable Spacings (ICLR 2024 · similarity 0.85)
- SpikingSSMs: Learning Long Sequences with Sparse and Parallel Spiking State Space Models (AAAI 2025 · similarity 0.85)
- Towards More Discriminative Feature Learning in SNNs with Temporal-Self-Erasing Supervision (AAAI 2025 · similarity 0.84)
- Efficient and Effective Time-Series Forecasting with Spiking Neural Networks (ICML 2024 · similarity 0.84)
- Unveiling the Spatial-temporal Effective Receptive Fields of Spiking Neural Networks (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
