基于同家族架构指导的大语言模型驱动神经网络生成：区分迁移与适应
Curating Same-Family Neural Networks for LLM-Guided Model Improvement: A Controlled Case Study
浏览论文内容
中文总结 AI 辅助
研究利用同家族强源模型改进弱目标模型，提出源引导候选生成协议，在CIFAR - 10和SVHN AlexNet等数据集上实验，结果表明该协议能显著提高准确性，且大语言模型是适应性改进而非复制，家族级分析有积极信号。
中文摘要 AI 辅助
大语言模型可生成神经网络修改，但无限制生成常无效或有害。本文研究在神经网络数据库中使用更强的同家族源模型改进弱目标模型这一更窄设定。提出源引导候选生成协议，含非源控制、源条件候选及无大语言模型消融。该协议分别报告有效性与准确性，仅在改进目标时选最佳有效候选。在CIFAR - 10和SVHN AlexNet上实验表明，源引导候选准确性显著高于非源候选，且大语言模型是适应而非复制源方法。家族级分析显示AlexNet和alt_nn1有明显积极信号。
英文摘要
Neural-network repositories contain executable models, recipes, input transformations, and measured accuracies. We study whether one same-family experiment can be curated as prompt guidance for LLM-based improvement of a low-performing target under equal generation and evaluation budgets. TuneNNGen extends NNGPT with a source-guided route and compares it with target-only generation on one CIFAR-10 target, two fixed source-selection rules, three code LLMs, and an additional SVHN target. Under the historical one-epoch search protocol, best-of-budget accuracy on the available evaluation split rises from 23.98% to 50.49% on CIFAR-10 and from 22.54% to 78.80% on SVHN. Selected five-epoch, three-seed means on train-derived validation splits retain gains of 40.94 points on CIFAR-10, 18.83 on Imagenette, and 7.27 on CIFAR-100. Direct-copy and negative-control analyses show that gains depend on source-target compatibility and LLM adaptation; stored source accuracy alone does not predict transferability.
发表机构
- University of Würzburg（维尔茨堡大学）
- Computer Vision Lab, CAIDAS（计算机视觉实验室，CAIDAS）
机构由 AI 辅助整理，请以论文原文为准。
