arXiv subjects

Huaiming Li

Publications and source records attributed to Huaiming Li.

3 recordsLinked to original sources

arXiv2026

Machine Learning-based Separation of the He I 10830{\AA} Chromospheric Signal: Quantitative Analysis of Chromosphere-Corona Intensity in the Quiet Sun

The He I 10830{\AA} line, a crucial optically thin chromospheric line, is frequently used to study coronal heating and vertical coupling across the chromosphere-corona interface. However, its images are severely contaminated by the strong photospheric background signal, hindering the analysis of fine chromospheric structures. Given the morphological differences between the Active Region (AR) and the Quiet Sun (QS), we proposed separating the He I 10830{\AA} chromospheric signal using two deep learning CNN models. Our model utilizes TiO images and cross-band learning to infer the He I 10830{\AA} photospheric background. The output is combined with an exponential absorption model to achieve quantitative analysis of the pure chromospheric component. Joint analysis of Solar Dynamics Observatory (SDO) data and the separated QS structures reveals a strong spatial negative correlation between chromospheric He I 10830{\AA} intensities(R approx -0.84 in 304{\AA} ), and significant layered coupling with EUV (171, 193, and 304{\AA}) radiation. Furthermore, strong He I 10830{\AA} absorption areas are highly correlated with regions of strong magnetic fields, while 171{\AA} radiative enhancement areas extend to the strong magnetic field edges and the mixed-polarity regions. These findings quantify the radiation intensity relationship between He I 10830{\AA} and EUV bands in the Quiet Sun. It also demonstrates the differences in heating characteristics between unipolar and mixed-polarity magnetic fields.

astro-ph.SR↗

arXiv2026

A Blueprint for Self-Evolving Coding Agents in Vehicle Aerodynamic Drag Prediction

High-fidelity vehicle drag evaluation is constrained less by solver runtime than by workflow friction: geometry cleanup, meshing retries, queue contention, and reproducibility failures across teams. We present a contract-centric blueprint for self-evolving coding agents that discover executable surrogate pipelines for predicting drag coefficient $C_d$ under industrial constraints. The method formulates surrogate discovery as constrained optimization over programs, not static model instances, and combines Famou-Agent-style evaluator feedback with population-based island evolution, structured mutations (data, model, loss, and split policies), and multi-objective selection balancing ranking quality, stability, and cost. A hard evaluation contract enforces leakage prevention, deterministic replay, multi-seed robustness, and resource budgets before any candidate is admitted. Across eight anonymized evolutionary operators, the best system reaches a Combined Score of 0.9335 with sign-accuracy 0.9180, while trajectory and ablation analyses show that adaptive sampling and island migration are primary drivers of convergence quality. The deployment model is explicitly ``screen-and-escalate'': surrogates provide high-throughput ranking for design exploration, but low-confidence or out-of-distribution cases are automatically escalated to high-fidelity CFD. The resulting contribution is an auditable, reusable workflow for accelerating aerodynamic design iteration while preserving decision-grade reliability, governance traceability, and safety boundaries.

cs.AI↗

arXiv2025

The FM Agent

Large language models (LLMs) are catalyzing the development of autonomous AI research agents for scientific and engineering discovery. We present FM Agent, a novel and general-purpose multi-agent framework that leverages a synergistic combination of LLM-based reasoning and large-scale evolutionary search to address complex real-world challenges. The core of FM Agent integrates several key innovations: 1) a cold-start initialization phase incorporating expert guidance, 2) a novel evolutionary sampling strategy for iterative optimization, 3) domain-specific evaluators that combine correctness, effectiveness, and LLM-supervised feedback, and 4) a distributed, asynchronous execution infrastructure built on Ray. Demonstrating broad applicability, our system has been evaluated across diverse domains, including operations research, machine learning, GPU kernel optimization, and classical mathematical problems. FM Agent reaches state-of-the-art results autonomously, without human interpretation or tuning -- 1976.3 on ALE-Bench (+5.2\%), 43.56\% on MLE-Bench (+4.0pp), up to 20x speedups on KernelBench, and establishes new state-of-the-art(SOTA) results on several classical mathematical problems. Beyond academic benchmarks, FM Agent shows considerable promise for both large-scale enterprise R\&D workflows and fundamental scientific research, where it can accelerate innovation, automate complex discovery processes, and deliver substantial engineering and scientific advances with broader societal impact.

cs.AI↗