22 April 2026 in IEEE Internet of Things Journal
With the global proliferation of electric vehicles (EVs), battery state monitoring has emerged as a foundation for ensuring operational safety and reliability. However, in real-world applications, battery systems operate within complex dynamic spatiotemporal contexts, characterized by nonstationary aging drift in the temporal dimension and cross-scale heterogeneity in the spatial dimension. While existing reviews extensively cover specific parameter estimation, most studies rely on static assumptions or isolated technical paths, failing to capture the deep spatiotemporal coupling mechanisms or meet the privacy and bandwidth constraints of distributed vehicular environments. To bridge this gap, this article provides a comprehensive review of battery monitoring algorithms, organized by temporal, spatial, and spatiotemporal dimensions. Furthermore, to address the limitations of centralized paradigms in managing heterogeneous data, we analyze a decoupling-interaction (DI) framework enabled by federated learning (FL). By shifting the monitoring paradigm toward cloud–edge collaborative intelligence, this framework facilitates localized spatiotemporal decoupling for heterogeneous data adaptation and cloud-based interaction for global evolutionary pattern mining. Finally, we offer insights into the open challenges of practical deployment, and further discuss the transition of monitoring algorithms toward semantic reasoning via foundation models, physics-constrained communication, and generative continual learning. This provides a systematic research perspective and future direction for battery state monitoring within dynamic spatiotemporal contexts.
5 March 2025 in The Canadian Journal of Chemical Engineering
16 August 2026 in SSRN Electronic Journal
Fine-tuning methods that combine federated learning with low-rank adaptation (LoRA) enable efficient collaborative training of large language models under resource constraints. Existing approaches widely adopt single-LoRA or multi-LoRA architectures. However, non-IID client data contain both global and personalized knowledge. Single-LoRA architectures tend to entangle them, while, multi-LoRA architectures lack persistent constraints between the global and private branches, causing the progressive discrepancy between local knowledge and aggregation knowledge. To address these limitations, this paper proposes FedPOD-LoRA, a federated fine-tuning method based on proximal optimization and dual-LoRA collaboration. First, we introduce a targeted proximal constraint on the global LoRA matrix B at the client side to reduce representation divergence between LoRA branches, thereby mitigating knowledge drift and fusion conflicts under non-IID data. Second, we further optimize the aggregated global Bmatrix using public data at the server side to enhance its global knowledge representation capability. Compared with five state-of-the-art (SOTA) methods, FedPOD-LoRA achieves better understanding performance, faster convergence. Across five natural language understanding tasks, including MNLI, SST-2, QNLI, QQP, and RTE, it improves accuracy by up to 2.5% and reduces the communication rounds required to reach the target performance on RTE and QNLI by 25.6% and 63.9%.
24 June 2026 in SSRN Electronic Journal
The efficacy of Large Language Model (LLM) agents is often constrained by the inefficiency of context utilization during runtime. Traditional ”log-centric” frameworks rely on syntax-heavy state representations like nested JSON, which impose a syntactic tax that diverts finite attention from semantic reasoning to structural parsing, leading to context dilution in long-horizon tasks. We propose Fat-Cat, a document-driven metacognitive architecture that transitions agent state management from passive logging to active semantic orchestration. Fat-Cat introduces two synergistic mechanisms: (1) Document-Driven State Modeling. By utilizing a markdown-based global workspace aligned with pre-training priors, we achieve representational alignment that collapses context growth from a quadratic 𝑂(𝑇 2) trajectory into a linear 𝑂(𝑇 ) one. (2) Closed-Loop Semantic Watcher. An independent auditor performs pre-commit verification to detect invalid state transitions and prevent error cascades. Additionally, a textual strategy evolution module facilitates parameter-free learning through methodological distillation. Extensive benchmarking suggests that document-driven state orchestration improves the reliability and context efficiency of LLM agents across multiple backbone models, especially in long-horizon reasoning and domain-grounded verification tasks.
10 April 2026 in SSRN Electronic Journal
The efficacy of Large Language Model (LLM) agents is often constrained by the inefficiency of context utilization during runtime. Traditional “log-centric” frameworks rely on syntax-heavy state representations like nested JSON, which impose a syntactic tax that diverts finite attention from semantic reasoning to structural parsing, leading to context dilution in long-horizon tasks. We propose Fat-Cat, a document-driven metacognitive architecture that transitions agent state management from passive logging to active semantic orchestration. Fat-Cat introduces two synergistic mechanisms: (1) Document-Driven State Modeling. By utilizing a Markdown-based semantic file system aligned with pre-training priors, we achieve representational alignment that collapses context growth from a quadratic 𝑂(𝑇 ²) trajectory into a linear 𝑂(𝑇 ) one. (2) Closed-Loop Semantic Watcher. An independent auditor performs pre-emptive audits, functioning as a semantic firewall to intercept hallucinations and prevent error cascades. Additionally, a textual strategy evolution module facilitates parameter-free learning through methodological distillation. Extensive benchmarking demonstrates that architectural-level representational alignment effectively enables open-weight models to rival or surpass proprietary systems, providing a robust framework for efficient autonomous agents.