Iterative LLM-NAS is equivalent to a parametric cross-entropy method with proven monotonic quality improvement, geometric convergence of elite probability, and a closed-form proxy reliability rho_S = (6/pi) arcsin(rho_P(SNR)/2), partially confirmed on 3300 architectures.
Lemur neural net- work dataset: Towards seamless automl
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
abstract
Neural networks are the backbone of modern artificial intelligence, but designing, evaluating, and comparing them remains labor-intensive. While numerous datasets exist for training, there are few standardized collections of the models themselves. We introduce LEMUR, an open-source dataset and framework that provides a large collection of PyTorch-based neural networks across tasks such as classification, segmentation, detection, and natural language processing. Each model follows a unified template, with configurations and results stored in a structured database to ensure consistency and reproducibility. LEMUR integrates automated hyperparameter optimization via Optuna, includes statistical analysis and visualization tools, and offers an API for seamless access to performance data. The framework is extensible, allowing researchers to add new models, datasets, or metrics without breaking compatibility. By standardizing implementations and unifying evaluation, LEMUR aims to accelerate AutoML research, enable fair benchmarking, and reduce barriers to large-scale neural network experimentation. To support adoption and collaboration, LEMUR and its plugins are released under the MIT license at: https://github.com/ABrain-One/nn-dataset https://github.com/ABrain-One/nn-plots https://github.com/ABrain-One/nn-vr
representative citing papers
Fine-tuned 7B LLMs generating unified diffs for neural architecture refinement achieve 66-75% valid rates and 64-66% mean first-epoch accuracy, outperforming full-generation baselines by large margins while cutting output length by 75-85%.
Closed-loop LLM search with AST-generated examples discovers non-standard channel widths that improve vision model performance over initial architectures on CIFAR-100.
Three-example few-shot prompting optimizes LLM-generated vision architectures while a whitespace-normalized hash provides 100x faster duplicate detection than AST parsing across seven benchmarks.
NN-RAG extracts 1,289 candidate neural modules from 19 PyTorch repositories, validates 941 of them, and supplies roughly 72% of the novel structures in the LEMUR dataset while enabling cross-repository migration.
Fine-tuned LLMs reach 80% accuracy predicting which dataset a neural network code performs better on, outperforming metadata prompts at 70%.
A 1.96M-parameter LiteDenoiseNet student model achieves 37.58 dB PSNR on full-resolution real image denoising benchmarks while running in 34-46 ms on mobile NPUs by leveraging NPU-compatible primitives and high-alpha knowledge distillation.
FractalNet automatically generates and tests over 1,200 CNN architectures based on recursive fractal templates, achieving up to 80.18% accuracy on CIFAR-10 after five training epochs.
MobileAgeNet uses a MobileNetV3-Large backbone with a regression head to achieve 4.65 years mean absolute error in age estimation and 14.4 ms on-device latency with 3.23 million parameters.
citing papers explorer
-
Convergence Theory for Iterative LLM-Based Neural Architecture Search: A Parametric Cross-Entropy Framework with Closed-Form Proxy Reliability
Iterative LLM-NAS is equivalent to a parametric cross-entropy method with proven monotonic quality improvement, geometric convergence of elite probability, and a closed-form proxy reliability rho_S = (6/pi) arcsin(rho_P(SNR)/2), partially confirmed on 3300 architectures.
-
Delta-Based Neural Architecture Search: LLM Fine-Tuning via Code Diffs
Fine-tuned 7B LLMs generating unified diffs for neural architecture refinement achieve 66-75% valid rates and 64-66% mean first-epoch accuracy, outperforming full-generation baselines by large margins while cutting output length by 75-85%.
-
Closed-Loop LLM Discovery of Non-Standard Channel Priors in Vision Models
Closed-loop LLM search with AST-generated examples discovers non-standard channel widths that improve vision model performance over initial architectures on CIFAR-100.
-
From Code to Prediction: Fine-Tuning LLMs for Neural Network Performance Classification in NNGPT
Fine-tuned LLMs reach 80% accuracy predicting which dataset a neural network code performs better on, outperforming metadata prompts at 70%.
-
Real Image Denoising with Knowledge Distillation for High-Performance Mobile NPUs
A 1.96M-parameter LiteDenoiseNet student model achieves 37.58 dB PSNR on full-resolution real image denoising benchmarks while running in 34-46 ms on mobile NPUs by leveraging NPU-compatible primitives and high-alpha knowledge distillation.
-
MobileAgeNet: Lightweight Facial Age Estimation for Mobile Deployment
MobileAgeNet uses a MobileNetV3-Large backbone with a regression head to achieve 4.65 years mean absolute error in age estimation and 14.4 ms on-device latency with 3.23 million parameters.