Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: What is the impact of varying the number of federated learning rounds on the model performance (accuracy, F1-score) and communication efficiency (throughput, bandwidth usage) when training on N-BaIoT. In this…

Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How does the model accuracy of federated learning-based malware detection compare to centralized training on the N-BaIoT dataset when evaluated using precision, recall, and F1-score metrics. This work…

Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How do different client sampling strategies (e.g., random, stratified, adaptive) affect the trade-off between communication efficiency and model accuracy in federated malware detection systems with. Personalized…

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: What is the impact of heterogeneous client data distributions on the generalization performance of federated deep neural networks for malware classification, and how can model personalization. In federated…

Abstract: This report synthesises findings from 8 peer-reviewed papers addressing the following research question: How does the F1-score of federated malware detection models trained on N-BaIoT transfer to unseen IoT network traffic datasets (e.g., BoT-IoT, CIC-IoT-2021) under varying differential privacy noise. This work…

Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: What is the impact of federated model aggregation frequency (e.g., 10, 50, 100 rounds) on the inference throughput for malware detection on resource-constrained IoT devices using the N-BaIoT dataset. This work…

Abstract: This report synthesises findings from 7 peer-reviewed papers addressing the following research question: How does the federated learning model's F1-score degradation compare to centralized training when using differential privacy noise levels of \$varepsilon\$=0.1, \$varepsilon\$=1, and \$varepsilon\$=10 on the…

Abstract: This report synthesises findings from 4 peer-reviewed papers addressing the following research question: How does federated transfer learning scale with varying numbers of IoT device clients in terms of communication efficiency and detection accuracy trade-offs, as measured by convergence speed and F1. Federated…

Abstract: This report synthesises findings from 8 peer-reviewed papers addressing the following research question: What is the impact of different federated learning aggregation strategies (e.g., FedAvg, FedProx) on malware detection performance across heterogeneous IoT device types when measured by. In this paper, we…

Abstract: This report synthesises findings from 7 peer-reviewed papers addressing the following research question: How does the performance of supervised versus unsupervised federated learning models compare on the N-BaIoT dataset using F1-score and AUC-ROC as evaluation metrics for cross-device malware detection. This work…

Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How does the bandwidth utilization of federated learning-based malware detection models scale with increasing device heterogeneity in IoT networks when employing model compression techniques (e.g.,. Federated…

Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: What is the impact of varying client participation rates and network heterogeneity on detection accuracy and model convergence in federated malware detection systems, measured through federated. This work…

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: How does federated transfer learning performance on N-BaIoT compare to centralized transfer learning in terms of cross-domain accuracy and F1 scores when evaluated on other IoT malware datasets. This work…

Abstract: This report synthesises findings from 7 peer-reviewed papers addressing the following research question: What is the impact of different federated learning aggregation techniques (e.g., FedAvg, FedProx) on the robustness of malware detection models in IoT networks with device heterogeneity, evaluated. This work…

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: How does the integration of quantum-enhanced federated learning affect the accuracy and convergence rate of malware detection models compared to classical federated learning in heterogeneous IoT. This work…

Abstract: This report synthesises findings from 13 peer-reviewed papers addressing the following research question: What is the inference efficiency (tokens/sec) of domain-adapted Baichuan-2 models on the FactCC benchmark when scaled to different batch sizes. Programming robots is complicated due to the lack of `plug-and-play'…

Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: How does cross-domain fine-tuning of Baichuan-2 on legal vs. biomedical text compare in terms of FactCC benchmark scores when evaluated with varying TruthfulQA misalignment thresholds. In the era of digital…

Abstract: This report synthesises findings from 16 peer-reviewed papers addressing the following research question: How does the token misalignment threshold parameter scale with model size across different LLM families when measuring hallucination rates on TruthfulQA. Multilingual large-scale Pretrained Language Models (PLMs)…

Abstract: This report synthesises findings from 10 peer-reviewed papers addressing the following research question: How do small language models for hallucination detection generalize across different LLM architectures when evaluated on TruthfulQA with varying token misalignment thresholds. Small Language Models (SLMs) offer…

Abstract: This report synthesises findings from 20 peer-reviewed papers addressing the following research question: How does quantized inference affect task performance on low-resource vision-language benchmarks. Multimodal Large Language Models (MLLM), which integrate large language models (LLMs) with vision models, aim to…

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: How do Baichuan 2 and Vicuna-13B compare on cross-lingual image captioning tasks in low-resource settings. In NLP, Zero-Shot Classification (ZSC) has become essential for enabling models to classify text into…

Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: Does the alignment of cross-domain fine-tuned LLMs with human preferences (e.g., via RLHF) influence their inference efficiency on the DS-1000 benchmark, as measured by tokens per second throughput. Fine-grained…

Abstract: This report synthesises findings from 4 peer-reviewed papers addressing the following research question: What is the impact of context window size on BigCodeBench pass@1 accuracy degradation across Llama 2 and Code Llama Python variants during cross-library API usage scenarios. Transformer language models typically…

Abstract: This report synthesises findings from 14 peer-reviewed papers addressing the following research question: What is the impact of mixing C/C++ and Python code samples in the training set on the F1 score for security vulnerability detection tasks in Codestral-7B versus Llama3-70B. Large Language Models (LLMs) have…

Abstract: This report synthesises findings from 11 peer-reviewed papers addressing the following research question: How robust are fine-tuned CodeT5 models when evaluated on cross-domain code migration tasks (e.g., Python to Java) using Pass@K and Exact Match metrics. Context: In the fast-paced evolution of software…