Resource-Efficient Iterative LLM-Based NAS with Feedback Memory

• URL: http://arxiv.org/abs/2603.12091v1
• Date: Thu, 12 Mar 2026 16:00:22 GMT
• Title: Resource-Efficient Iterative LLM-Based NAS with Feedback Memory
• Abstract summary: Neural Architecture Search (NAS) automates network design, but conventional methods demand substantial computational resources.<n>We propose a closed-loop pipeline leveraging large language models (LLMs) to iteratively generate, evaluate, and refine convolutional neural network architectures.
• Score: 49.44875022114861
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Neural Architecture Search (NAS) automates network design, but conventional methods demand substantial computational resources. We propose a closed-loop pipeline leveraging large language models (LLMs) to iteratively generate, evaluate, and refine convolutional neural network architectures for image classification on a single consumer-grade GPU without LLM fine-tuning. Central to our approach is a historical feedback memory inspired by Markov chains: a sliding window of $K{=}5$ recent improvement attempts keeps context size constant while providing sufficient signal for iterative learning. Unlike prior LLM optimizers that discard failure trajectories, each history entry is a structured diagnostic triple -- recording the identified problem, suggested modification, and resulting outcome -- treating code execution failures as first-class learning signals. A dual-LLM specialization reduces per-call cognitive load: a Code Generator produces executable PyTorch architectures while a Prompt Improver handles diagnostic reasoning. Since both the LLM and architecture training share limited VRAM, the search implicitly favors compact, hardware-efficient models suited to edge deployment. We evaluate three frozen instruction-tuned LLMs (${\leq}7$B parameters) across up to 2000 iterations in an unconstrained open code space, using one-epoch proxy accuracy on CIFAR-10, CIFAR-100, and ImageNette as a fast ranking signal. On CIFAR-10, DeepSeek-Coder-6.7B improves from 28.2% to 69.2%, Qwen2.5-7B from 50.0% to 71.5%, and GLM-5 from 43.2% to 62.0%. A full 2000-iteration search completes in ${\approx}18$ GPU hours on a single RTX~4090, establishing a low-budget, reproducible, and hardware-aware paradigm for LLM-driven NAS without cloud infrastructure.

Related papers

• Consensus-gated Multi-Agent Neural Architecture Search for Seismic Fault Segmentation [0.17188280334580194]
  Neural networks for seismic fault segmentation are often borrowed from computer vision and medical imaging domains.<n>Classical neural architecture search (NAS) is restricted to hand-crafted search spaces and large compute budgets.<n>We present an agentic NAS system in which a panel of three large language models (Claude, GPT-5.1, and Gemini2.5Pro) debates each candidate architecture to unanimous consensus.
  arXiv  Detail & Related papers  (2026-08-14T02:41:37Z)
• Delta-Based Neural Architecture Search: LLM Fine-Tuning via Code Diffs [48.83701310501069]
  Large language models (LLMs) show strong potential for neural architecture generation.<n>Existing approaches produce complete model implementations from scratch.<n>We propose Delta-Code Generation, where fine-tuned LLMs generate compact unified diffs.
  arXiv  Detail & Related papers  (2026-05-06T13:32:05Z)
• Xe-Forge: Multi-Stage LLM-Powered Kernel Optimization for Intel GPU [38.81440160993858]
  Porting deep learning algorithms to new hardware accelerators requires developers to apply the same low-level optimizations to every Triton kernel in their code-base.<n>We present Xe-Forge, a multi-stage LLM-powered pipeline that automates this process for Intel GPU.<n>A curated knowledge base encodes Intel GPU constraints that are absent from LLM training data, keeping the model within architecturally valid bounds.
  arXiv  Detail & Related papers  (2026-04-16T20:21:01Z)
• $\nabla$-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Latent Space [71.23672814629448]
  $nabla$-Reasoner is an iterative generation framework that integrates differentiable optimization over token logits into the decoding loop.<n>$nabla$-Reasoner achieves over 20% accuracy improvement on a challenging mathematical reasoning benchmark.
  arXiv  Detail & Related papers  (2026-03-05T08:42:54Z)
• Beyond GEMM-Centric NPUs: Enabling Efficient Diffusion LLM Sampling [14.471123653746275]
  Diffusion Large Language Models (dLLMs) introduce iterative denoising to enable parallel token generation.<n>Our design employs lightweight non-GEMM vector primitives, in-place memory reuse strategies, and a decoupled mixed-precision memory hierarchy.
  arXiv  Detail & Related papers  (2026-01-28T15:37:50Z)
• NNGPT: Rethinking AutoML with Large Language Models [36.90850535125572]
  NNGPT is an open-source framework that turns a large language model (LLM) into a self-improving AutoML engine for neural network development.<n>It integrates within one unified workflow five synergistic LLM-based pipelines: zero-shot architecture synthesis, hyper parameter optimization, code-aware accuracy/early-stop prediction, and reinforcement learning.<n>The system has already generated over 5K validated models, proving NNGPT as an autonomous AutoML engine.
  arXiv  Detail & Related papers  (2025-11-25T14:10:44Z)
• dInfer: An Efficient Inference Framework for Diffusion Language Models [54.80918957287927]
  Diffusion-based large language models (dLLMs) have emerged as a promising alternative to autoregressive (AR) LLMs.<n>We present dInfer, an efficient and framework for dLLM inference.
  arXiv  Detail & Related papers  (2025-10-09T16:19:42Z)
• Enhancing LLM-based Fault Localization with a Functionality-Aware Retrieval-Augmented Generation Framework [14.287359838639608]
  FaR-Loc is a framework that enhances method-level fault localization.<n> FaR-Loc consists of three key components: LLM Functionality Extraction, Semantic Retrieval, and LLM Re-ranking.<n>Our experiments on the widely used Defects4J benchmark show that FaR-Loc outperforms state-of-the-art LLM-based baselines.
  arXiv  Detail & Related papers  (2025-09-24T20:37:11Z)
• PentaRAG: Large-Scale Intelligent Knowledge Retrieval for Enterprise LLM Applications [5.4838799162708245]
  We present PentaRAG, a five-layer module that routes each query through two instant caches.<n>We show that PentaRAG cuts average GPU time to 0.248 seconds per query, roughly half that of a naive RAG baseline.<n>Results demonstrate that a layered routing strategy can deliver freshness, speed, and efficiency simultaneously in production-grade RAG systems.
  arXiv  Detail & Related papers  (2025-06-18T07:54:53Z)
• Search for Efficient Large Language Models [52.98684997131108]
  Large Language Models (LLMs) have long held sway in the realms of artificial intelligence research. Weight pruning, quantization, and distillation have been embraced to compress LLMs, targeting memory reduction and inference acceleration. Most model compression techniques concentrate on weight optimization, overlooking the exploration of optimal architectures.
  arXiv  Detail & Related papers  (2024-09-25T21:32:12Z)
• RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval [24.472784635757016]
  RetrievalAttention is a training-free approach to both accelerate attention computation and reduce GPU memory consumption.<n>We show that RetrievalAttention achieves near full attention accuracy while only requiring access to 1--3% of the data.
  arXiv  Detail & Related papers  (2024-09-16T17:59:52Z)
• Hardware-Aware Parallel Prompt Decoding for Memory-Efficient Acceleration of LLM Inference [23.633481089469836]
  Auto-regressive decoding of Large Language Models (LLMs) results in significant overheads in their hardware performance.<n>We propose a novel parallel prompt decoding that requires only $0.0002$% trainable parameters, enabling efficient training on a single A100-40GB GPU in just 16 hours.<n>Our approach demonstrates up to 2.49$times$ speedup and maintains a minimal memory overhead of just $0.0004$%.
  arXiv  Detail & Related papers  (2024-05-28T22:19:30Z)
• DISTFLASHATTN: Distributed Memory-efficient Attention for Long-context LLMs Training [82.06732962485754]
  FlashAttention effectively reduces the quadratic peak memory usage to linear in training transformer-based large language models (LLMs) on a single GPU. We introduce DISTFLASHATTN, a memory-efficient attention mechanism optimized for long-context LLMs training. It achieves 1.67x and 1.26 - 1.88x speedup compared to recent Ring Attention and DeepSpeed-Ulysses.
  arXiv  Detail & Related papers  (2023-10-05T03:47:57Z)

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.