RevoNAD: Reflective Evolutionary Exploration for Neural Architecture Design

• URL: http://arxiv.org/abs/2512.05403v1
• Date: Fri, 05 Dec 2025 03:47:50 GMT
• Title: RevoNAD: Reflective Evolutionary Exploration for Neural Architecture Design
• Abstract summary: RevoNAD is a reflective evolutionary orchestrator that bridges LLM-based reasoning with feedback-aligned architectural search.<n>It achieves state-of-the-art performance across CIFAR10, CIFAR100, ImageNet16-120, COCO-5K, and Cityscape.
• Score: 13.808891421521432
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent progress in leveraging large language models (LLMs) has enabled Neural Architecture Design (NAD) systems to generate new architecture not limited from manually predefined search space. Nevertheless, LLM-driven generation remains challenging: the token-level design loop is discrete and non-differentiable, preventing feedback from smoothly guiding architectural improvement. These methods, in turn, commonly suffer from mode collapse into redundant structures or drift toward infeasible designs when constructive reasoning is not well grounded. We introduce RevoNAD, a reflective evolutionary orchestrator that effectively bridges LLM-based reasoning with feedback-aligned architectural search. First, RevoNAD presents a Multi-round Multi-expert Consensus to transfer isolated design rules into meaningful architectural clues. Then, Adaptive Reflective Exploration adjusts the degree of exploration leveraging reward variance; it explores when feedback is uncertain and refines when stability is reached. Finally, Pareto-guided Evolutionary Selection effectively promotes architectures that jointly optimize accuracy, efficiency, latency, confidence, and structural diversity. Across CIFAR10, CIFAR100, ImageNet16-120, COCO-5K, and Cityscape, RevoNAD achieves state-of-the-art performance. Ablation and transfer studies further validate the effectiveness of RevoNAD in allowing practically reliable, and deployable neural architecture design.

Related papers

• A Robust Point Cloud Analysis Framework Inspired By Primary Visual Cortex [78.18514741653475]
  We propose a Dendritic-Connected Continuous-Coupled Neural Network (DC-CCNN) architecture for point cloud analysis.<n>By combining discrete and continuous encoding, our design replaces traditional Multilayer Perceptrons with more efficient and robust BINNs.<n>We introduce a Neuro-Inspired Robust Modulation-and-Readout Module (NRMR) to enhance feature stability and decision robustness.
  arXiv  Detail & Related papers  (2026-06-12T09:17:52Z)
• Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction [52.32112533846212]
  MRI reconstruction is an inherently ill-posed inverse problem, since incomplete measurements admit many plausible solutions.<n>We address this limitation by moving reconstruction to discrete multi-scale latent space and posing it as autoregressive next-acceleration-scale prediction.<n>We show that our approach delivers improved reconstruction performance across diverse sampling patterns under extreme undersampling.
  arXiv  Detail & Related papers  (2026-05-19T04:40:50Z)
• Architecture-Aware Multi-Design Generation for Repository-Level Feature Addition [53.50448142467294]
  RAIM is a multi-design and architecture-aware framework for repository-level feature addition.<n>It shifts away from linear patching by generating multiple diverse implementation designs.<n>Experiments on the NoCode-bench Verified dataset demonstrate that RAIM establishes a new state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-03-02T12:50:40Z)
• ReNF: Rethinking the Design Space of Neural Long-Term Time Series Forecasters [48.79331759671512]
  We introduce a Multiple Neural Forecasting Theorem that provides a theoretical basis for our approach.<n>We propose Boosted Direct Output (BDO), a novel forecasting strategy that combines the advantages of both Auto-Regressive (AR) and Direct Output (DO)
  arXiv  Detail & Related papers  (2025-09-30T08:05:59Z)
• InvFussion: Bridging Supervised and Zero-shot Diffusion for Inverse Problems [76.39776789410088]
  This work introduces a framework that combines the strong performance of supervised approaches and the flexibility of zero-shot methods.<n>A novel architectural design seamlessly integrates the degradation operator directly into the denoiser.<n> Experimental results on the FFHQ and ImageNet datasets demonstrate state-of-the-art posterior-sampling performance.
  arXiv  Detail & Related papers  (2025-04-02T12:40:57Z)
• NADER: Neural Architecture Design via Multi-Agent Collaboration [37.48197934228379]
  We introduce NADER, a novel framework that formulates neural architecture design (NAD) as a multi-agent collaboration problem.<n>We propose the Reflector, which effectively learns from immediate feedback and long-term experiences.<n>Unlike previous LLM-based methods that use code to represent neural architectures, we utilize a graph-based representation.
  arXiv  Detail & Related papers  (2024-12-26T13:07:03Z)
• An Architectural Approach to Enhance Deep Long-Tailed Learning [5.214135587370722]
  We introduce Long-Tailed Differential Architecture Search (LTDAS)<n>We conduct extensive experiments to explore architectural components that demonstrate better performance on long-tailed data.<n>This ensures that the architecture obtained through our search process incorporates superior components.
  arXiv  Detail & Related papers  (2024-11-09T07:19:56Z)
• Decoupled Sequence and Structure Generation for Realistic Antibody Design [45.72237864940556]
  A dominant paradigm is to train a model to jointly generate the antibody sequence and the structure as a candidate.<n>We propose an antibody sequence-structure decoupling (ASSD) framework, which separates sequence generation and structure prediction.<n>ASSD shows improved performance in various antibody design experiments, while the composition-based objective successfully mitigates token repetition of non-autoregressive models.
  arXiv  Detail & Related papers  (2024-02-08T13:02:05Z)
• Towards Architecture-Agnostic Untrained Network Priors for Image Reconstruction with Frequency Regularization [14.73423587548693]
  We propose efficient architecture-agnostic techniques to directly modulate the spectral bias of network priors. We show that, with just a few lines of code, we can reduce overfitting in underperforming architectures and close performance gaps with high-performing counterparts. Results signify for the first time that architectural biases, overfitting, and runtime issues of untrained network priors can be simultaneously addressed without architectural modifications.
  arXiv  Detail & Related papers  (2023-12-15T18:01:47Z)
• RING-NeRF : Rethinking Inductive Biases for Versatile and Efficient Neural Fields [1.1816466088976698]
  We propose the RING-NeRF architecture which includes two inductive biases. A single reconstruction process takes advantage of those inductive biases and experimentally demonstrates on-par performances. We also design a single reconstruction process that takes advantage of those inductive biases and experimentally demonstrates on-par performances.
  arXiv  Detail & Related papers  (2023-12-06T08:54:04Z)
• Dynamically Grown Generative Adversarial Networks [111.43128389995341]
  We propose a method to dynamically grow a GAN during training, optimizing the network architecture and its parameters together with automation. The method embeds architecture search techniques as an interleaving step with gradient-based training to periodically seek the optimal architecture-growing strategy for the generator and discriminator.
  arXiv  Detail & Related papers  (2021-06-16T01:25:51Z)
• STONNE: A Detailed Architectural Simulator for Flexible Neural Network Accelerators [5.326345912766044]
  STONNE is a cycle-accurate, highly-modular and highly-extensible simulation framework. We show how it can closely approach the performance results of the publicly available BSV-coded MAERI implementation.
  arXiv  Detail & Related papers  (2020-06-10T19:20:52Z)
• Disturbance-immune Weight Sharing for Neural Architecture Search [96.93812980299428]
  We propose a disturbance-immune update strategy for model updating. We theoretically analyze the effectiveness of our strategy in alleviating the performance disturbance risk.
  arXiv  Detail & Related papers  (2020-03-29T17:54:49Z)
• Stabilizing Differentiable Architecture Search via Perturbation-based Regularization [99.81980366552408]
  We find that the precipitous validation loss landscape, which leads to a dramatic performance drop when distilling the final architecture, is an essential factor that causes instability. We propose a perturbation-based regularization - SmoothDARTS (SDARTS) - to smooth the loss landscape and improve the generalizability of DARTS-based methods.
  arXiv  Detail & Related papers  (2020-02-12T23:46:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.