Issue Ebook
cover story
Current Issue
lssue 4, 2026
Editors-in-Chief
Yunhe Pan, Aiguo Fei
Advisor Xicheng Lu
ISSN 3069-8928
CN 33-1439/TP
IF (2024 JCR) 2.9
| Submit Now |
- Current Issue
- Cover Articles
- Archive
- Virtual Issues
- Online First
Volume 27 Issue 4，2026
Regular Papers;Review
Low-altitude UAV swarm ISAC: new opportunities and challenges Enhanced Publication AI Introduction
Abstract：With the rapid development of the low-altitude economy, low-altitude unmanned aerial vehicle (UAV) swarms are emerging as important components of sixth-generation (6G) mobile communication networks, facilitating “full coverage” and “Internet of Intelligence.” Integrated sensing and communication (ISAC) deeply integrates sensing functionality into wireless communication networks by sharing wireless infrastructures and resources such as base stations, antennas, radio frequency chains, and signal waveforms, thereby significantly improving the performance of low-altitude UAV swarms. This paper reviews the research status of low-altitude UAV swarm ISAC systems, analyzes the new challenges arising from key features of UAV swarms, including low–slow–small characteristics, high density, large quantity, complex low-altitude environments, and high swarm coordination requirements, presents a vision for future deployment, and proposes the so-called “Ten Ones” performance metrics tailored to low-altitude UAV swarm ISAC. To realize these ambitious key performance indicators for future UAV swarm ISAC, several promising technologies are discussed, such as new array architectures, including extremely-large multiple-input multiple-output (XL-MIMO), sparse XL-MIMO, and reconfigurable antenna arrays, sparse time–frequency resource allocation, and channel knowledge maps. Furthermore, the potential of exploiting UAV swarms as airborne ISAC platforms is discussed. Finally, future research directions are outlined, offering a guideline for the design and development of low-altitude UAV swarm ISAC systems.Keywords：Unmanned aerial vehicle (UAV) swarm;Integrated sensing and communication (ISAC);Sparse extremely-large multiple-input multiple-output (XL-MIMO);Reconfigurable antenna arrays;Sparse time–frequency resource allocation;Channel knowledge maps (CKMs)23|9|0Updated：2026-04-24Regular Papers;Research Article
Three-dimensional affordance segmentation for object point cloud driven by language instructions AI Introduction
Abstract：The location where a robot grasps an object is closely related to the task type. For the same object, different user requirements may necessitate different grasping strategies. Visual affordance serves as a reliable source of prior knowledge for manipulation. Existing methods learn affordance from images or videos, but planar affordance lacks the spatial information required for 6-degree-of-freedom (6-DoF) manipulation. Furthermore, current approaches are limited to affordances associated with predefined categories and cannot directly infer affordances from user instructions. To address such limitations, we propose a novel task: instruction-driven three-dimensional (3D) object affordance segmentation. To support this research, we introduce an instruction–affordance dataset (IAD), a challenging dataset consisting of 7190 object instances across 20 common object categories, paired with 624 manipulation instructions that specify the corresponding affordances. To evaluate generalization to novel commands, our dataset includes both seen and unseen settings. Building on this, we design an instruction-driven 3D affordance segmentation (IDAS) network, which extracts point cloud features and integrates instruction features layer by layer. Given a user instruction, our method segments suggested manipulation regions on the object’s point cloud, thereby guiding the selection of optimal grasp poses. Experimental results show that our method outperforms other related approaches under both seen and unseen settings, demonstrating generalization ability to diverse user commands and unknown affordances.Keywords：Visual affordance;Point cloud segmentation;Open vocabulary;Multimodal fusion;Service robot19|10|0Updated：2026-04-24Regular Papers;Research Article
Image fragment carving based on DCT semantics and an adjustment factor Enhanced Publication AI Introduction
Abstract：The recovery of evidence from fragmented image files is a prominent research focus in the field of file carving. To address image fragment reassembly, this paper analyzes the Joint Photographic Experts Group (JPEG) image structure and proposes a fragment connection weighting algorithm based on discrete cosine transform (DCT) semantic features, along with a weight adjustment factor that leverages image compression characteristics. By integrating these components, the algorithm effectively determines the fragment sequence in JPEG files, and a practical carving algorithm is designed. Experiments conducted on disk and memory demonstrate that the adjustment factor-based algorithm outperforms the DCT-only method in identifying true best matches (reducing false positives). Disk experiments achieve an average carving precision of 94.4%, surpassing existing methods, while memory experiments validate the feasibility of the approach, along with a theoretical analysis of failure scenarios caused by interference from software such as Windows Photo Viewer.Keywords：Discrete cosine transform;Fragmented image files;Memory media12|12|0Updated：2026-04-24Regular Papers;Research Article
Reversible data hiding in encrypted domain based on NTRU and Chinese remainder theorem AI Introduction
Abstract：Reversible data hiding in the encrypted domain (RDH-ED) based on homomorphic encryption provides a promising approach for privacy-preserving data sharing, yet existing methods based on the Nth-degree truncated polynomial ring unit (NTRU) face a fundamental conflict between embedding capacity and reversibility, often requiring preprocessing of plaintext, which in turn compromises randomness of the ciphertext obtained. To address these issues, a novel RDH-ED scheme integrating the Chinese remainder theorem (CRT) with the NTRU cryptosystem is proposed in this study. The proposed scheme operates without any preprocessing of the plaintext and constructs multichannel redundancy in the ciphertext domain, thereby fully preserving the original polynomial structure of the plaintext. By employing a CRT-based encoding, multiple bits of information are enabled to be carried by a single polynomial coefficient, achieving an embedding capacity of 503 bits per polynomial with moderate-sized parameters. Moreover, the embedded data can be extracted before decryption via pre-negotiated coprime parameters, offering greater operational flexibility. Rigorous mathematical constraints ensure that the redundancy term is automatically eliminated during decryption, thereby guaranteeing lossless recovery of the original content. Experimental results demonstrate that the proposed scheme achieves a substantially higher embedding capacity compared to predominant RDH-ED methods based on NTRU, Paillier, and ElGamal cryptosystems, without compromising security or efficiency.Keywords：Reversible data hiding;NTRU cryptosystem;Chinese remainder theorem (CRT);Multichannel redundancy14|10|0Updated：2026-04-24Regular Papers;Research Article
WSC optimizer: an optimization tool for wafer-scale chip architecture exploration Enhanced Publication AI Introduction
Abstract：In recent years, mature advanced packaging technologies have increasingly enabled the integration of multiple small dies into larger chips, while retaining chip-scale density and high-bandwidth interconnects. To address the inefficiencies of manual design and the challenges of heterogeneous optimization in wafer-scale chip (WSC) development, we systematically explore key factors in WSC architecture design. We integrate chip layout, operator mapping, and hardware–software co-design, and formulate the WSC architecture exploration problem as a multi-objective optimization task. First, we establish a hierarchical architecture model for WSCs, unifying the quantification of core constraints and interconnect topology constraints; second, we propose a hierarchical multi-objective collaborative optimization framework to jointly optimize physical constraints and task mapping communication patterns; finally, we develop a WSC optimizer toolchain that supports mixed-granularity simulation and generates optimal configurations for representative workloads. Experimental results demonstrate that compared with traditional computer architectures, the optimized architectures generated by our WSC optimizer achieve up to a 22× throughput improvement and a 5× latency reduction in application domains, such as cryptographic decryption and signal processing.Keywords：Wafer-scale chip;Hardware–software co-design;Chip layout;Design space exploration15|11|0Updated：2026-04-24Regular Papers;Research Article
RetryTrigger: intelligent inference duplication for enhancing LLM resilience to hardware transient faults Enhanced Publication AI Introduction
Abstract：Large language models (LLMs) have exhibited outstanding performance across a wide range of natural language processing (NLP) tasks. However, the rising prevalence of hardware transient faults has made silent data corruptions (SDCs) in LLMs increasingly problematic, severely degrading output quality and user experience. State-of-the-art protection schemes primarily rely on hardware-assisted algorithm-based fault tolerance (ABFT) or boundary-setting-driven online fault tolerance (FT2) for selective layers, yet these solutions suffer from strict hardware dependencies, substantial overhead, or incomplete coverage. To address these limitations, we propose RetryTrigger, a novel hardware-free fault-aware inference methodology capable of handling all potential faults. During LLM inference, RetryTrigger dynamically collects runtime output features (e.g., maximum probability, top-k probability gaps, output entropy, logits statistics, and inference latency), which are used to train a LightGBM meta-model. This meta-model accurately predicts whether duplicate inference should be performed, thereby effectively mitigating faults while preserving efficiency without additional hardware dependence. Extensive experiments on seven representative LLMs (including T5-Small, RoBERTa, BioMedBERT, Qwen2.5-Coder-0.5B/7B, MiniMind, and Opt) demonstrate that RetryTrigger reduces SDC rates by up to 95.33% (on average 92.97%) and achieves a minimal performance overhead of 2.4012% (on average 4.1167%), offering a superior balance between reliability and efficiency compared to state-of-the-art solutions.Keywords：Large language models;System resilience;Intelligent fault detection;Inference duplication;Transient faults26|10|0Updated：2026-04-24
SEE MORE
0 1 Hybrid-augmented intelligence: collaboration and cognition
0 4 Applications of artificial intelligence in intelligent manufacturing: a review*
0 5 Challenges and opportunities: from big data to knowledge in AI 2.0
0 6 Current trends in the development of intelligent unmanned autonomous systems
0 7 Recent advances in efficient computation of deep convolutional neural networks
0 8 Terahertz time-domain spectroscopy and micro-cavity components for probing samples: a review
0 9 Cross-media analysis and reasoning: advances and directions
SEE MORE
Videos
-
00:02:512023 Issue 1 | Scalability and efficiency challenges for the exascale supercomputing system: practice of a parallel supporting environment on the Sunway exascale prototype system
2023-12-30Play Total： 30 - 00:02:30
2023 Issue 6 | Model division multiple access for semantic communications
2023-12-30Play Total： 22 - 00:02:15
2022 Issue 10 | Discussion on a new paradigm of endogenous security towards 6G networks
2023-12-30Play Total： 11 - 00:02:22
2022 Issue 12 | Technology trends in large-scale high-efficiency network computing
2023-12-30Play Total： 8 - 00:02:48
2022 Issue 6 | Self-deployed execution environment for high performance computing
2022-08-03Play Total： 17 - 00:02:24
2022 Issue 2 | A full-process intelligent trial system for smart court
2022-05-17Play Total： 20 - 00:02:37
2022 Issue 3 | Automatic protocol reverse engineering for industrial control systems with dynamic taint analysis
2022-05-17Play Total： 10 - 00:05:36
P1 Speech by Academician Baoyan Duan
2022-04-17Play Total： 24 - 00:02:27
P2 Speech by Professor Min Sheng, Xidian University
2022-04-17Play Total： 17 - 00:02:37
P3 Speech by Professor Yunsong Li, Xidian University
2022-04-17Play Total： 20
SEE MORE
- Address：Zhejiang University Press, 148 Tianmushan Road, Hangzhou, China Postal code：310028
- Tel：+86-571-88273162 Email：fitee@zju.edu.cn
- Copyright © 2000 - 2025, Zhejiang University Press 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0