mlsanityData to 14 Oct 2026

MONAQ: Multi-Objective Neural Architecture Querying for Time-Series Analysis on Resource-Constrained Devices

arXiv 2505.10607 ↗First posted 15 May 2025cs.LG cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

The growing use of smartphones and IoT devices necessitates efficient time-series analysis on resource-constrained hardware, which is critical for sensing applications such as human activity recognition and air quality prediction. Recent efforts in hardware-aware neural architecture search (NAS) automate architecture discovery for specific platforms; however, none focus on general time-series analysis with edge deployment. Leveraging the problem-solving and reasoning capabilities of large language models (LLM), we propose MONAQ, a novel framework that reformulates NAS into Multi-Objective Neural Architecture Querying tasks. MONAQ is equipped with multimodal query generation for processing multimodal time-series inputs and hardware constraints, alongside an LLM agent-based multi-objective search to achieve deployment-ready models via code generation. By integrating numerical data, time-series images, and textual descriptions, MONAQ improves an LLM's understanding of time-series data. Experiments on fifteen datasets demonstrate that MONAQ-discovered models outperform both handcrafted models and NAS baselines while being more efficient.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 30 of its 53 references that are in the map. First posted on arXiv 15 May 2025.

Builds on

30 of 53 references.

References · 30

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,476
  3. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,439
  4. 2018ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardwarecited by 673
  5. 2018The UEA multivariate time series classification archive, 2018cited by 338
  6. 2020MCUNet: Tiny Deep Learning on IoT Devicescited by 273
  7. 2022On-Device Training Under 256KB Memorycited by 160
  8. 2023Neural Architecture Search: Insights from 1000 Paperscited by 139
  9. 2017Hello Edge: Keyword Spotting on Microcontrollerscited by 110
  10. 2021MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learningcited by 89
  11. 2024Language Models Still Struggle to Zero-shot Reason about Time Seriescited by 88
  12. 2023LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimizationcited by 87
18 more
  1. 2024AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoMLcited by 87
  2. 2021HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmarkcited by 84
  3. 2023Time Series as Images: Vision Transformer for Irregularly Sampled Time Seriescited by 79
  4. 2021A Comprehensive Survey on Hardware-Aware Neural Architecture Searchcited by 61
  5. 2024Agents in Software Engineering: Survey, Landscape, and Visioncited by 60
  6. 2019Best Practices for Scientific Research on Neural Architecture Searchcited by 54
  7. 2024TimeSeriesExam: A time series understanding examcited by 49
  8. 2024Evaluating Large Language Models on Time Series Feature Understanding: A Comprehensive Taxonomy and Benchmarkcited by 38
  9. 2024A Picture is Worth A Thousand Numbers: Enabling LLMs Reason about Time Series via Visualizationcited by 33
  10. 2024Universal Time-Series Representation Learning: A Surveycited by 30
  11. 2023LLM Performance Predictors are good initializers for Architecture Searchcited by 29
  12. 2024See it, Think it, Sorted: Large Multimodal Models are Few-shot Time Series Anomaly Analyzerscited by 20
  13. 2020Small-Footprint Keyword Spotting with Multi-Scale Temporal Convolutioncited by 15
  14. 2024LeMo-NADe: Multi-Parameter Neural Architecture Discovery with LLMscited by 11
  15. 2023MicroNAS: Memory and Latency Constrained Hardware-Aware Neural Architecture Search for Time Series Classification on Microcontrollerscited by 8
  16. 2023Heterogeneous Graph Neural Architecture Search with GPT-4cited by 7
  17. 2024Sensor-Aware Classifiers for Energy-Efficient Time Series Applications on IoT Devicescited by 2
  18. 2024Optimizing TinyML: The Impact of Reduced Data Acquisition Rates for Time Series Classification on Microcontrollerscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025Multi-View Encoders for Performance Prediction in LLM-Based Agentic Workflowscited by 3
  2. 2026GenAutoML: An Agentic Framework for Dynamic Architecture Generation and Optimization in Time-Series Analysiscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026TimeART: Towards Agentic Time Series Reasoning via Tool-Augmentation
  2. 2024LLaMA-NAS: Efficient Neural Architecture Search for Large Language Models
  3. 2025TimeSeriesGym: A Scalable Benchmark for (Time Series) Machine Learning Engineering Agents
  4. 2026InA-Probe: Instruction-Aware Active Probing for Time Series Forecasting with LLMs
  5. 2025Training-Free Time Series Classification via In-Context Reasoning with LLM Agents
  6. 2024Evolution and Efficiency in Neural Architecture Search: Bridging the Gap Between Expert Design and Automated Optimization
  7. 2025When LLM Meets Time Series: Can LLMs Perform Multi-Step Time Series Reasoning and Inference
  8. 2025TS-Agent: Understanding and Reasoning Over Raw Time Series via Iterative Insight Gathering