Alignment plays a crucial role in Large Language Models (LLMs) in aligning with human preferences on a specific task/domain. Traditional alignment methods suffer from catastrophic forgetting, where models lose previously learned values when adapting to new preferences or domains. We introduce LifeAlign, a novel framework for lifelong alignment that enables LLMs to maintain consistent human preference alignment across sequential learning tasks without forgetting previously learned values. Our approach consists of two key innovations. First, we propose a focalized preference optimization strategy that aligns LLMs with new preferences while preventing the erosion of alignment acquired from previous tasks. Second, we develop a short-to-long memory consolidation mechanism that merges denoised short-term preference representations into stable long-term memory using intrinsic dimensionality reduction, enabling efficient storage and retrieval of alignment patterns across diverse domains. We evaluate LifeAlign across multiple sequential alignment tasks spanning different domains and preference types. Experimental results demonstrate that our method achieves superior performance in maintaining both preference alignment quality and knowledge retention compared to existing lifelong learning approaches.

Automatic math correction aims to check students’ solutions to mathematical problems via artificial intelligence technologies. Most existing studies focus on judging the final answer at the problem level, while they ignore detailed feedback on each step in a math problem-solving process, which requires abilities of semantic understanding and reasoning. In this paper, we propose a reinforcement learning (RL)-based method to boost large language model (LLM) for step-level automatic math correction, named StepAMC. Particularly, we convert the step-level automatic math correction within the text classification task into an RL problem to enhance the reasoning capabilities of LLMs. Then, we design a space-constrained policy network to improve the stability of RL. Then, we introduce a fine-grained reward network to convert the binary human feedback into a continuous value. We conduct extensive experiments over two benchmark datasets and the results show that our model outperforms the eleven strong baselines.

With the rapid development of geographic location technology and the explosive growth of data, a large amount of spatial data is outsourced to the cloud server for reducing the local high storage and computing burdens, but at the same time causes security issues. Thus, extensive privacy-preserving spatial data query schemes have been proposed. Most of the existing schemes use Asymmetric Scalar-Product-Preserving Encryption (ASPE) to encrypt data, but ASPE has proven to be insecure against known plaintext attack. And the existing schemes require users to provide more information about query range and thus generate a large amount of ciphertexts, which causes high storage and computational burdens. To solve these issues, based on enhanced ASPE designed in our conference version, we first propose a basic Privacy-preserving Spatial Data Query (PSDQ) scheme by using a new unified index structure, which only requires users to provide less information about query range. Then, we propose an enhanced PSDQ scheme (PSDQ$^+$) by using Geohash-based$R$-tree structure (called$GR$-tree) and efficient pruning strategy, which greatly reduces the query time. Formal security analysis proves that our schemes achieve Indistinguishability under Chosen Plaintext Attack (IND-CPA), and extensive experiments demonstrate that our schemes are efficient in practice.

With the rapid development of Location-Based Services (LBS), a large number of LBS providers outsource spatial data to cloud servers to reduce their high computational and storage burdens, but meanwhile incur some security issues such as location privacy leakage. Thus, extensive privacy-preserving LBS schemes have been proposed. However, the existing solutions using Bloom filter do not take into account the redundant bits that do not map information in Bloom filter, resulting in high computational overheads, and reveal the inclusion relationship in Bloom filter. To solve these issues, we propose an efficient Privacy-preserving Spatial Range Query (PSRQ) scheme by skillfully combining Geohash algorithm with Circular Shift and Coalesce Bloom Filter (CSC-BF) framework and Symmetric-key Hidden Vector Encryption (SHVE), which not only greatly reduces the computational cost of generating token but also speeds up the query efficiency on large-scale datasets. In addition, we design a Confused Bloom Filter (CBF) to confuse the inclusion relationship by confusing the values of 0 and 1 in the Bloom filter. Base on this, we further propose a more secure and practical enhanced scheme PSRQ+by using CBF and Geohash algorithm, which can support more query ranges and achieve adaptive security. Finally, formal security analysis proves that our schemes are secure against Indistinguishability under Chosen-Plaintext Attacks (IND-CPA) and PSRQ+achieves adaptive IND-CPA, and extensive experimental tests demonstrate that our schemes using million-level dataset improve the query efficiency by 100x compared with previous state-of-the-art solutions.

With the rapid development of Intelligent Transportation System (ITS), a large number of spatial data are generated in ITS. Although outsourcing spatial data to the cloud server can reduce the high local computation and storage overheads, it will also lead to security and privacy issues. Therefore, it is necessary to have a survey to specifically summarize these advanced privacy-preserving spatial data query schemes. However, the existing surveys considering both location information and keywords of spatial data only summarize the spatial keyword query scheme in plaintext environment, they do not consider the privacy of spatial data. Although there are some surveys on privacy-preserving spatial data query, they only focus on the location information of spatial data without considering descriptive keywords. Therefore, to understand the progress and research trends in the field, we give a comprehensive survey on secure spatial data query in ITS to summarize and analyze the most advanced solutions. Then, we make a comprehensive and detailed comparison of existing solutions in terms of query function, index structure, time complexity, security, etc. Finally, we show some open challenges and potential research directions for privacy-preserving spatial data query.

With the rapid development of geographic location technology and the explosive growth of data, a large amount of spatio-textual data is outsourced to the cloud server to reduce the local high storage and computing burdens, but at the same time causes security issues such as data privacy leakage. Thus, extensive privacy-preserving spatial keyword query schemes have been proposed. Most of the existing schemes use Asymmetric Scalar-Product-Preserving Encryption (ASPE) for encryption, but ASPE has proven to be insecure. And the existing spatial range query schemes require users to provide more information about the query range and generate a large amount of ciphertext, which causes high storage and computational burdens. To solve these issues, in this paper we introduce some random numbers and a random permutation to enhance the security of ASPE scheme, and then propose a novel privacy-preserving Spatial Keyword Query (SKQ) scheme based on the enhanced ASPE and Geohash algorithm. In addition, we design a more Lightweight Spatial Keyword Query (LSKQ) scheme by using a unified index for spatial range and multiple keywords, which not only greatly decreases SKQ’s storage and computational costs but also requires users to provide little information about query region. Finally, formal security analysis proves that our schemes have Indistinguishability under Chosen Plaintext Attack (IND-CPA), and extensive experiments demonstrate that our enhanced scheme is efficient and practical.

Cloud-assisted Industrial Internet of Things (IIoT) systems are increasingly deployed in various applications such as location-based services. Outsourcing data to cloud servers can help minimize local data storage and computation overheads, but it may introduce security and privacy concerns. Therefore, privacy-preserving spatial keyword search has been extensively explored in the literature. However, existing solutions still reveal the order of the spatio-textual similarity values between the query point and all data objects, and do not support searching for arbitrary geometric regions. To solve these issues, in this article we propose a privacy-preserving threshold spatial keyword search (TSKS) scheme. Specifically, we use the polynomial fitting technology, vector space model, and randomizable matrix multiplication technology to allow the cloud server to find relevant objects that are within some arbitrary geometric range and contain all query keywords. Finally, formal security analysis proves that our scheme can protect the privacy of data sets and queries, and extensive experiments demonstrate that our scheme is efficient and practical.