Closed-source large language models deliver strong performance but have limited downstream customizability. Semi-open models, combining both closed-source and public layers, were introduced to improve customizability. However, parameters in the closed-source layers are found vulnerable to recovery attacks. In this paper, we explore the design of semi-open models with fewer closed-source layers, aiming to increase customizability while ensuring resilience to recovery attacks. We analyze the contribution of closed-source layer to the overall resilience and theoretically prove that in a deep transformer-based model, there exists a transition layer such that even small recovery errors in layers before this layer can lead to recovery failure. SCARA employs a fine-tuning-free metric to estimate the maximum number of layers that can be publicly accessible for customization. We apply it to five models (1.3B to 70B parameters) to construct semi-open models, validating their customizability on six downstream tasks and assessing their resilience against various recovery attacks on sixteen benchmarks. We compare SCARA to baselines and observe that it generally improves downstream customization performance and offers similar resilience with over 10 times fewer closed-source parameters. We empirically investigate the existence of transition layers, analyze the effectiveness of our scheme and finally discuss its limitations. Open-sourcing more parameters and structure details apparently enhances downstream customizability. However, Zanella-Beguelin et al. (2021) showed that semi-open LLMs with only a few closed-source parameters are vulnerable to model recovery attacks. Recovery attackers query the closed-source module and then train a new module that imitates its functionality. This can lead to the full replication and theft of closed-source modules (Solaiman, 2023). Recovery attackers targeting fully closed-source models seek to fine-tune a new model that precisely replicates the closed-source model (Tamber et al., 2024; Dubiński et al., 2024). In contrast, attackers in semi-open settings are not required to exactly replicate the closed-source module. Instead, they can fine-tune the closed-source module alongside the public module to reconstruct the overall functionality. While open-sourcing more layers enhances downstream flexibility, it also facilitates easier replication.

Existing works have established multiple benchmarks to highlight the security risks associated with Code GenAI. These risks are primarily reflected in two areas: a model's potential to generate insecure code (insecure coding) and its utility in cyberattacks (cyberattack helpfulness). While these benchmarks have made significant strides, there remain opportunities for further improvement. For instance, many current benchmarks tend to focus more on a model's ability to provide attack suggestions rather than its capacity to generate executable attacks. Additionally, most benchmarks rely heavily on static evaluation metrics (e.g., LLM judgment), which may not be as precise as dynamic metrics such as passing test cases. Furthermore, some large-scale benchmarks, while efficiently generated through automated methods, could benefit from more expert verification to ensure data quality and relevance to security scenarios. Conversely, expert-verified benchmarks, while offering high-quality data, often operate at a smaller scale. For insecure code, we introduce a new methodology for data creation that combines experts with automatic generation. Our methodology ensures the data quality while enabling large-scale generation. We also associate samples with test cases to conduct code-related dynamic evaluation. For cyberattack helpfulness, we set up a real environment and construct samples to prompt a model to generate actual attacks, along with dynamic metrics in our environment. Furthermore, it better identifies the security risks of SOTA models in insecure coding and cyberattack helpfulness. PLT to the SOTA code agent, Cursor, and, for the first time, identify non-trivial security risks in this advanced coding agent. Code GenAI, including specific code generation models and general large language models, have shown remarkable capabilities in code generation (Austin et al., 2021; Chen et al., 2021; DeepSeek, 2022; Dong et al., 2023; Hui et al., 2024), reasoning (Gu et al., 2024), and debugging (Tian et al., 2024). Together with these exciting new capabilities comes concern over these models' security risks. Recent research (Bhatt et al., 2023; Pearce et al., 2022) showed that code GenAI can produce insecure code, which significantly hinders the real-world deployment of AI-generated code. Moreover, these models can also be weaponized to facilitate cyberattacks. To understand the security risks of code GenAI, existing works developed several benchmarks to evaluate a code generation model's risk in producing insecure or vulnerable code (insecure coding) (Bhatt et al., 2023; 2024), as well as its potential to facilitate cyberattacks (cyberattack helpfulness) (Bhatt et al., 2024; Yuan et al., 2024). However, as demonstrated in Table 1, these benchmarks are foundationally limited.

Text generation, a key component in applications such as dialogue systems, relies on decoding algorithms that sample strings from a language model distribution. Traditional methods, such as top-$k$ and top-$\pi$, apply local normalisation to the model's output distribution, which can distort it. In this paper, we investigate the effect of this distortion by introducing globally-normalised versions of these decoding methods. Additionally, we propose an independent Metropolis-Hastings algorithm to approximate sampling from globally-normalised distributions without explicitly computing them. Our empirical analysis compares the performance of local and global normalisation across two decoding algorithms (top-$k$ and top-$\pi$) with various hyperparameters, using Pythia language models. Results show that, in most configurations, global decoding performs worse than the local decoding version of the same algorithms -- despite preserving the distribution's integrity. Our results suggest that distortion is an important feature of local decoding algorithms.

The literature on concept formation has demonstrated that humans are capable of learning concepts incrementally, with a variety of attribute types, and in both supervised and unsupervised settings. Many models of concept formation focus on a subset of these characteristics, but none account for all of them. In this paper, we present TRESTLE, an incremental account of probabilistic concept formation in structured domains that unifies prior concept learning models. TRESTLE works by creating a hierarchical categorization tree that can be used to predict missing attribute values and cluster sets of examples into conceptually meaningful groups. It updates its knowledge by partially matching novel structures and sorting them into its categorization tree. Finally, the system supports mixed-data representations, including nominal, numeric, relational, and component attributes. We evaluate TRESTLE's performance on a supervised learning task and an unsupervised clustering task. For both tasks, we compare it to a nonincremental model and to human participants. We find that this new categorization model is competitive with the nonincremental approach and more closely approximates human behavior on both tasks. These results serve as an initial demonstration of TRESTLE's capabilities and show that, by taking key characteristics of human learning into account, it can better model behavior than approaches that ignore them.

The geopolitical division between the indigenous Chakma population and mainstream Bangladesh creates a significant cultural and linguistic gap, as the Chakma community, mostly residing in the hill tracts of Bangladesh, maintains distinct cultural traditions and language. Developing a Machine Translation (MT) model or Chakma to Bangla could play a crucial role in alleviating this cultural-linguistic divide. Thus, we have worked on MT between CCP-BN(Chakma-Bangla) by introducing a novel dataset of 15,021 parallel samples and 42,783 monolingual samples of the Chakma Language. Moreover, we introduce a small set for Benchmarking containing 600 parallel samples between Chakma, Bangla, and English. We ran traditional and state-of-the-art models in NLP on the training set, where fine-tuning BanglaT5 with back-translation using transliteration of Chakma achieved the highest BLEU score of 17.8 and 4.41 in CCP-BN and BN-CCP respectively on the Benchmark Dataset. As far as we know, this is the first-ever work on MT for the Chakma Language. Hopefully, this research will help to bridge the gap in linguistic resources and contribute to preserving endangered languages. Our dataset link and codes will be published soon.

Background: Conventional prediction methods such as logistic regression and gradient boosting have been widely utilized for disease onset prediction for their reliability and interpretability. Deep learning methods promise enhanced prediction performance by extracting complex patterns from clinical data, but face challenges like data sparsity and high dimensionality. Methods: This study compares conventional and deep learning approaches to predict lung cancer, dementia, and bipolar disorder using observational data from eleven databases from North America, Europe, and Asia. Models were developed using logistic regression, gradient boosting, ResNet, and Transformer, and validated both internally and externally across the data sources. Discrimination performance was assessed using AUROC, and calibration was evaluated using Eavg. Findings: Across 11 datasets, conventional methods generally outperformed deep learning methods in terms of discrimination performance, particularly during external validation, highlighting their better transportability. Learning curves suggest that deep learning models require substantially larger datasets to reach the same performance levels as conventional methods. Calibration performance was also better for conventional methods, with ResNet showing the poorest calibration. Interpretation: Despite the potential of deep learning models to capture complex patterns in structured observational healthcare data, conventional models remain highly competitive for disease onset prediction, especially in scenarios involving smaller datasets and if lengthy training times need to be avoided. The study underscores the need for future research focused on optimizing deep learning models to handle the sparsity, high dimensionality, and heterogeneity inherent in healthcare datasets, and find new strategies to exploit the full capabilities of deep learning methods.

This study proposes a novel artificial intelligence (AI) driven flight computer, integrating an online free-retraining-prediction model, a swarm control, and an obstacle avoidance strategy, to track dynamic targets using a distributed drone swarm for military applications. To enable dynamic target tracking the swarm requires a trajectory prediction capability to achieve intercept allowing for the tracking of rapid maneuvers and movements while maintaining efficient path planning. Traditional predicative methods such as curve fitting or Long ShortTerm Memory (LSTM) have low robustness and struggle with dynamic target tracking in the short term due to slow convergence of single agent-based trajectory prediction and often require extensive offline training or tuning to be effective. Consequently, this paper introduces a novel robust adaptive bidirectional fuzzy brain emotional learning prediction (BFBEL-P) methodology to address these challenges. The controller integrates a fuzzy interface, a neural network enabling rapid adaption, predictive capability and multi-agent solving enabling multiple solutions to be aggregated to achieve rapid convergence times and high accuracy in both the short and long term. This was verified through the use of numerical simulations seeing complex trajectory being predicted and tracked by a swarm of drones. These simulations show improved adaptability and accuracy to state of the art methods in the short term and strong results over long time domains, enabling accurate swarm target tracking and predictive capability.

Cloth-changing person re-identification (CC-ReID) poses a significant challenge in computer vision. A prevailing approach is to prompt models to concentrate on causal attributes, like facial features and hairstyles, rather than confounding elements such as clothing appearance. Traditional methods to achieve this involve integrating multi-modality data or employing manually annotated clothing labels, which tend to complicate the model and require extensive human effort. In our study, we demonstrate that simply reducing feature correlations during training can significantly enhance the baseline model's performance. We theoretically elucidate this effect and introduce a novel regularization technique based on density ratio estimation. This technique aims to minimize feature correlation in the training process of cloth-changing ReID baselines. Our approach is model-independent, offering broad enhancements without needing additional data or labels. We validate our method through comprehensive experiments on prevalent CC-ReID datasets, showing its effectiveness in improving baseline models' generalization capabilities.

The Internet of Things (IoT) is one of the fastest-growing computing industries. By the end of 2027, more than 29 billion devices are expected to be connected. These smart devices can communicate with each other with and without human intervention. This rapid growth has led to the emergence of new types of malware. However, traditional malware detection methods, such as signature-based and heuristic-based techniques, are becoming increasingly ineffective against these new types of malware. Therefore, it has become indispensable to find practical solutions for detecting IoT malware. Machine Learning (ML) and Deep Learning (DL) approaches have proven effective in dealing with these new IoT malware variants, exhibiting high detection rates. In this paper, we bridge the gap in research between the IoT malware analysis and the wide adoption of deep learning in tackling the problems in this domain. As such, we provide a comprehensive review on deep learning based malware analysis across various categories of the IoT domain (i.e. Extended Internet of Things (XIoT)), including Industrial IoT (IIoT), Internet of Medical Things (IoMT), Internet of Vehicles (IoV), and Internet of Battlefield Things (IoBT).

Spatio-temporal area-level datasets play a critical role in official statistics, providing valuable insights for policy-making and regional planning. Accurate modeling and forecasting of these datasets can be extremely useful for policymakers to develop informed strategies for future planning. Echo State Networks (ESNs) are efficient methods for capturing nonlinear temporal dynamics and generating forecasts. However, ESNs lack a direct mechanism to account for the neighborhood structure inherent in area-level data. Ignoring these spatial relationships can significantly compromise the accuracy and utility of forecasts. In this paper, we incorporate approximate graph spectral filters at the input stage of the ESN, thereby improving forecast accuracy while preserving the model's computational efficiency during training. We demonstrate the effectiveness of our approach using Eurostat's tourism occupancy dataset and show how it can support more informed decision-making in policy and planning contexts.