Surface erosion of the coaxial thermocouple probe initiates continuous bridging of thermoelectric materials on the insulation layer surface, forming new temperature measurement junctions. This inherent ability to measure continuous self-erosion ensures the operational reliability of the coaxial thermocouples in high-temperature ablative environments. However, the fabrication of a high-temperature electrical insulation layer and a high-adhesion insulating layer in the coaxial thermocouples remains a challenge. Inspired by calcium carbonate/oxalate crystals in jujube leaves that strengthen the leaves, a bioinspired structural ceramic (BSC) mimicking these needle-like crystals is designed. This BSC demonstrates excellent high-temperature insulation (with insulation impedance of 2.55 kΩ at 1210 °C) and adhesion strength (35.3 Newtons). The BSC is successfully used as the insulating layer in a K-type coaxial thermocouple. The generation rules for surface junctions are systematically studied, revealing that stable and reliable measurement junctions can be created when the sandpaper grit does not exceed 600#. Static test results show that the K-type coaxial thermocouple ranges from 200 °C to 1200 °C with an accuracy of 1.1%, a drift rate better than 0.0137%/h, and hysteresis better than 0.81%. Dynamic test results show that the response time is 1.08 ms. The K-type coaxial thermocouple can withstand a high-temperature flame impact for 300 s at 1200 °C, as well as over forty cycles of high-power laser thermal shock, while maintaining good response characteristics. Therefore, the K-type coaxial thermocouple designed in this study provides an ideal solution for long-term temperature monitoring of the thermal components of aerospace engines under extremely high-temperature, high-speed, and strong thermal shock conditions.

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.

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.