Skip to main content
QUICK REVIEW

[论文解读] Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning

Xiangyu Yin, Ming Du|arXiv (Cornell University)|Mar 5, 2026
Advanced X-ray Imaging Techniques被引用 0
一句话总结

Ptychi-Evolve 使用大语言模型自动发现并演化用于镊光重建的正则化算法,在三个具有挑战性的数据集上优于基线,同时保留可解释的算法谱系。

ABSTRACT

Ptychography is a computational imaging technique widely used for high-resolution materials characterization, but high-quality reconstructions often require the use of regularization functions that largely remain manually designed. We introduce Ptychi-Evolve, an autonomous framework that uses large language models (LLMs) to discover and evolve novel regularization algorithms. The framework combines LLM-driven code generation with evolutionary mechanisms, including semantically-guided crossover and mutation. Experiments on three challenging datasets (X-ray integrated circuits, low-dose electron microscopy of apoferritin, and multislice imaging with crosstalk artifacts) demonstrate that discovered regularizers outperform conventional reconstructions, achieving up to +0.26 SSIM and +8.3~dB PSNR improvements. Besides, Ptychi-Evolve records algorithm lineage and evolution metadata, enabling interpretable and reproducible analysis of discovered regularizers.

研究动机与目标

  • 为病态镊光重建的正则化策略的自主发现提供动机。
  • 实现LLM驱动的可执行正则化器生成与演化,并与 Pty-Chi 集成。
  • 提供用于可重复发现的灵活评估与历史跟踪管线。
  • 在跨多样数据集的基线重建上展示改进。

提出的方法

  • LLM 引擎将可执行正则化器代码生成为与 Pty-Chi 兼容的 Python 函数。
  • 语义引导的交叉与变异推动正则化器的进化 refined。
  • 多模态评估(真实地 ground-truth、人类专家与视觉语言模型)指导选择。
  • 带元数据的历史管理实现可解释性与检查点的能力。
  • 安全性与输入验证确保生成的代码在受限环境中可安全执行。
  • 检查点与压缩保持在长时间运行中的发现可控。
Figure 1 : System architecture of Ptychi-Evolve showing the discovery loop. The LLM Engine generates regularizer code, which is executed by the Pty-Chi reconstruction library. The Evaluator assesses reconstruction quality and records metrics to the History Manager. Context from the history informs b
Figure 1 : System architecture of Ptychi-Evolve showing the discovery loop. The LLM Engine generates regularizer code, which is executed by the Pty-Chi reconstruction library. The Evaluator assesses reconstruction quality and records metrics to the History Manager. Context from the history informs b

实验结果

研究问题

  • RQ1自动化的LLM驱动系统是否能够发现超越人工设计的新颖正则化策略用于镊光?
  • RQ2语义引导的进化操作是否比随机重组产生更有效的正则化器?
  • RQ3所发现的正则化器在不同成像模态和伪影类型下是否稳健?
  • RQ4保持完整的算法谱系是否能够为何有效以及原因提供可解释的洞见?

主要发现

数据集算法成功率最佳Δ SSIMΔ PSNR持续时间
X-ray IC10083%0.785+0.26+8.3 dB16.5 小时
Apoferritin14797%0.836+0.12+3.2 dB29.5 小时
Multislice9794%0.871+0.18+8.0 dB10.5 小时
  • 发现的正则化器在三类数据集上显著优于基线的 LSQML 重建(SSIM 提升最高达 +0.26,PSNR 最高提升 +8.3 dB)。
  • X 射线 IC 显示最大的绝对改进,成功率为 83%,突出网格伪影的挑战。
  • apoferritin 与多层数据集在高成功率下获得显著提升(分别为 97% 和 94%)。
  • 表现最佳的多层正则化器整合四个梯度分量,包括 Charbonnier-TV、梯度排除、Gram 正交性以及带对比自适应权重的谱掩模,带来显著改进。
  • 该框架记录完整的算法谱系和进化元数据,使对有效正则化策略的可解释分析成为可能。
Figure 2 : Visual comparison of reconstructed phase images at representative iterations. Columns show a reference image, baseline reconstruction, and best discovered regularizer. For multislice and apoferritin, the reference is the simulated ground-truth phase; for X-ray IC, the reference is a long-
Figure 2 : Visual comparison of reconstructed phase images at representative iterations. Columns show a reference image, baseline reconstruction, and best discovered regularizer. For multislice and apoferritin, the reference is the simulated ground-truth phase; for X-ray IC, the reference is a long-

更好的研究,从现在开始

从论文设计到论文写作,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。