Recent works have demonstrated the effectiveness of retrieval augmentation in the Event Argument Extraction (EAE) task. However, existing retrieval-based EAE methods have two main limitations: (1) input length constraints and (2) the gap between the retriever and the inference model. These issues limit the diversity and quality of the retrieved information. In this paper, we propose a Compressive Memory-based Retrieval (CMR) mechanism for EAE, which addresses the two limitations mentioned above. Our compressive memory, designed as a dynamic matrix that effectively caches retrieved information and supports continuous updates, overcomes the limitations of input length. Additionally, after pre-loading all candidate demonstrations into the compressive memory, the model further retrieves and filters relevant information from the memory based on the input query, bridging the gap between the retriever and the inference model. Extensive experiments show that our method achieves new state-of-the-art performance on three public datasets (RAMS, WikiEvents, ACE05), significantly outperforming existing retrieval-based EAE methods.

Uncoordinated random-access protocols are attractive for underwater acoustic (UWA) networks due to their simplicity and low overhead, especially for data collection applications in scuba diving. However, the performance is limited by severe collisions and the challenging UWA channel, including rich multipath and time-varying channel (caused by Doppler effects and user movements). Existing UWA physical-layer waveforms struggle to resolve collisions while maintaining high data rates. This paper presents ZCMod, a Zadoff-Chu (ZC) sequence-based modulation that assigns unique ZC sequences to users to mitigate interference and encodes multiple bits via cyclic shifts for high data rates. To address UWA-specific challenges, ZCMod introduces two key designs: 1) shape-based demodulation, which tracks channel response shifts to combat multipath effects; 2) auxiliary modulation, where each symbol is modulated with two ZC sequences—one for channel estimation and the other for data transmission—to handle fast time-varying channels. Experiments and simulations demonstrate that a) ZCMod achieves more robust BER performance and eliminates error floors compared to state-of-the-art (SOTA) methods in slight time-varying channels; and b) ZCMod maintains stable throughput in fast time-varying channels, while SOTA approaches suffer significant degradation.

The combination of Spiking Neural Networks (SNNs) and Vision Transformers (ViTs) holds potential for achieving both energy efficiency and high performance, particularly suitable for edge vision applications. However, a significant performance gap still exists between SNN-based ViTs and their ANN counterparts. Here, we first analyze why SNN-based ViTs suffer from limited performance and identify a mismatch between the vanilla self-attention mechanism and spatio-temporal spike trains. This mismatch results in degraded spatial relevance and limited temporal interactions. To address these issues, we draw inspiration from biological saccadic attention mechanisms and introduce an innovative Saccadic Spike Self-Attention (SSSA) method. Specifically, in the spatial domain, SSSA employs a novel spike distribution-based method to effectively assess the relevance between Query and Key pairs in SNN-based ViTs. Temporally, SSSA employs a saccadic interaction module that dynamically focuses on selected visual areas at each timestep and significantly enhances whole scene understanding through temporal interactions. Building on the SSSA mechanism, we develop a SNN-based Vision Transformer (SNN-ViT). Extensive experiments across various visual tasks demonstrate that SNN-ViT achieves state-of-the-art performance with linear computational complexity. The effectiveness and efficiency of the SNN-ViT highlight its potential for power-critical edge vision applications.

With the development of informatization of IoT devices, nonterrestrial networks (NTNs) are becoming more and more important. NTN, including air and space networks, face challenges, such as high-computational complexity, bandwidth requirements, and memory constraints. An intelligent automatic modulation classification (AMC) mechanism based on neural networks plays a pivotal role in enhancing spectrum efficiency, throughput, and link reliability. Past work in AMC has evolved from likelihood-based and feature-based methods to traditional machine learning techniques and, more recently, to deep neural networks (DNNs). However, existing DNN architectures pose challenges for NTN due to high-computational complexity, bandwidth requirements, and memory consumption. Addressing this problems, we proposes a spiking transformer-based model for AMC, exploiting temporal dynamics for enhanced performance. Biologically inspired spiking neural networks enable us to exploit the sparse and binarized activation properties of spiking neurons, allowing us to build AMC models with high-energy efficiency and high availability that can be used in NTN systems. Furthermore, we introduce a weight binarization method to reduce the model size, which also further reduces the bandwidth and memory requirements of AMC in NTN edge deployment. Experimental results demonstrate the superiority of our approach over state-of-the-art methods, with the binarized model achieving comparable accuracy at a fraction of the size.

Uncoordinated random-access protocols are well-suited for underwater acoustic (UWA) networks due to their simplicity and low overhead. However, their performance is hindered by severe collisions and the challenging characteristics of UWA channels such as rich multipath and Doppler effect. Existing UWA physical layer waveforms struggle to resolve collisions while maintaining high data rates. This paper introduces ZCMod, a high-rate waveform allowing uncoordinated concurrent random access in UWA networks. ZCMod employs a Zadoff–Chu (ZC) sequence-based modulation that assigns unique ZC sequences to users to minimize inter-user interference and encodes multiple bits through cyclic shifts of the sequences to improve data rates. ZCMod further addresses the unique challenges of UWA channels via two new designs: 1) a shape-based demodulation approach that estimates the data-induced shift of channel response shape between the preamble and data symbols to handle rich multipath, and 2) an auxiliary modulation approach that modulates each data symbol with two ZC sequences, one for extracting current channel response shape and the other for data modulation, to handle the fast time-varying channel. Experimental results in a lake and a swimming pool and extensive simulation results show that a) ZCMod achieves around 100% higher throughput compared with the state-of-the-art (SOTA) approaches in quasi-static channels, and b) ZCMod maintains comparable throughput in fast time-varying channels as in quasi-static conditions, where the SOTA approaches experience significant degradation.

Linear chirp-based underwater acoustic communication has been widely used due to its reliability and long-range transmission capability. However, unlike the counterpart chirp technology in wireless - LoRa, its throughput is severely limited by the number of modulated chirps in a symbol. The fundamental challenge lies in the underwater multi-path channel, where the delayed signal may cause inter-symbol and intra-symbol interfere. In this paper, we present UWLoRa+, a system that realizes the same chirp modulation as LoRa with higher data rate, and address the multi-path challenge via the following new designs: a) we replace the linear chirp used by LoRa with the non-linear chirp to reduce the signal interference range and the collision probability; b) we design an algorithm that first demodulates each path and then combines the demodulation results of detected paths; and c) we replace the Hamming codes used by LoRa with the non-binary LDPC codes to mitigate the impact of the inevitable collision. Experiment results show that the new designs improve the bit error rate (BER) by 3 times, and the packet error rate (PER) significantly, compared with the LoRa's naive design. Compared with an state-of-the-art system for decoding underwater LoRa chirp signal, UWLoRa+ improves the throughput by up to 50 times.

Uncoordinated random-access protocols are especially suitable for underwater acoustic networks with long propagation delays due to their simplicity. However, their performance is limited by severe collisions caused by uncoordinated access, and the current modulations cannot handle the collisions under the multipath environment. In this paper, we propose a new modulation and a new demodulation algorithm to resolve collisions. In particular, we adopt a Zadoff-Chu (ZC) sequence with cyclic shifts as the modulation, and assign users with different ZC sequences to minimize inter-user interference. To combat the multipath challenge, we leverage the insight that the multipath interference pattern is almost constant within the same packet and the modulated data only shifts the pattern, and develop a pattern-based demodulation algorithm. Trace-driven simulation results show that our new approach allows at least five users, and outperforms the existing approach by at least 8dB. In the future, we intend to develop a real-time system in a realistic environment.