MIDAS is a differentiable neural architecture search method that replaces static architecture parameters with dynamic, input‑specific parameters generated through self‑attention. It localizes architecture selection to each spatial patch of the activation map and introduces a parameter‑free, topology‑aware search space that models two‑edge node topology. Evaluations on DARTS, NAS‑Bench‑201, and RDARTS show that MIDAS achieves state‑of‑the‑art accuracy on CIFAR‑10 and CIFAR‑100 and discovers optimal or near‑optimal architectures in benchmark spaces.
arXiv:2608. 14472v1 Announce Type: cross Abstract: Neural Architecture Search (NAS) aims to automate neural network architecture design, reducing reliance on human expertise.
ONNX-Net introduces a universal representation for neural architectures using natural language descriptions, enabling instant performance prediction across diverse search spaces. The authors present ONNX-Bench, a benchmark of over 600k architecture–accuracy pairs compiled from open‑source NAS‑bench networks in ONNX format. Experiments demonstrate strong zero‑shot predictive performance with minimal pretraining, overcoming the limitations of cell‑based, graph‑encoded approaches.
arXiv:2608. 14443v1 Announce Type: cross Abstract: Neural Architecture Search (NAS) is naturally formulated as a bilevel optimization problem, where the upper-level optimizes the architecture using validation performance and the lower-level trains network parameters using training loss.
DIFFNAS is a new neural architecture search framework that eliminates the need for performance predictors by using a discrete graph diffusion model conditioned on zero‑cost performance proxies such as MeCo and SWAP. The method achieves state‑of‑the‑art results on NAS‑Bench‑201 for CIFAR‑10 and CIFAR‑100, matching optimal architectures while cutting search cost by up to 17.5× compared to DiffusionNAG. Although its accuracy on the MobileNetV3 search space is lower than DiffusionNAG, the results demonstrate the potential of diffusion models combined with training‑free metrics for scalable architecture search.
arXiv:2607. 11826v1 Announce Type: cross Abstract: Neural Architecture Search (NAS) has automated the design of deep learning models but traditionally requires massive computational resources, often measured in thousands of GPU-days.
The paper introduces a semi-automated method for structuring neural architecture search (NAS) using large language models (LLMs). It defines a high-level template of architectural attributes, which an LLM populates by analyzing research papers to create a diverse search space. The authors then propose FairNAD, a multi-type mutation strategy that combines fair idea sampling, Pareto-aware mutation, LLM-driven iterative mutation, and a fine-grained feedback loop to efficiently explore this space, achieving significant performance gains on CIFAR-10, CIFAR-100, and ImageNet16-120.
arXiv:2403. 10318v3 Announce Type: replace Abstract: Recent advances have shifted the paradigm of tabular learning toward tabular foundation models, yet their accuracy relies on a heavy inference cost that scales poorly with context size.
arXiv:2606. 10294v1 Announce Type: cross Abstract: Deploying neural networks on unconventional hardware demands architectures that co-optimize task accuracy and platform-specific constraints such as energy cost, physical non-idealities, and numerical precision.
arXiv:2606. 07664v1 Announce Type: cross Abstract: Neuroevolution is a representative neural architecture search paradigm that evolves both network topology and weights through evolutionary algorithms.
arXiv:2609.14184v1 Announce Type: new Abstract: Neural Architecture Search (NAS) automates network design, but evaluating a single candidate requires training it to convergence, making exhaustive sea...
RiPPLE is a method for ranking neural architectures across an entire search space using only a small fraction of early training data. It treats partial training as labels for a limited set of anchor architectures, extrapolates their learning curves, and propagates these surrogate labels to other architectures without requiring per‑candidate features. The approach is evaluated on twelve benchmark cells from four search‑space families and the larger DARTS space, demonstrating its effectiveness in ranking quality, label efficiency, and architecture selection.
arXiv:2608. 14621v1 Announce Type: cross Abstract: Long-term memory is increasingly central to LLM agents, yet memory design remains a highly coupled architecture problem: what to encode, how to store it, how to retrieve it, and how to manage it can vary substantially across tasks and backbone models.