mlsanityData to 14 Oct 2026

A multi-agentic framework for real-time, autonomous freeform metasurface design

arXiv 2503.20479 ↗First posted 26 Mar 2025physics.app-ph cs.AI cs.MAcited by 14

Not on the map. Why?

Abstract (arXiv)

Innovation in nanophotonics currently relies on human experts who synergize specialized knowledge in photonics and coding with simulation and optimization algorithms, entailing design cycles that are time-consuming, computationally demanding, and frequently suboptimal. We introduce MetaChat, a multi-agentic design framework that can translate semantically described photonic design goals into high-performance, freeform device layouts in an automated, nearly real-time manner. Multi-step reasoning is enabled by our Agentic Iterative Monologue (AIM) paradigm, which coherently interfaces agents with code-based tools, other specialized agents, and human designers. Design acceleration is facilitated by Feature-wise Linear Modulation-conditioned Maxwell surrogate solvers that support the generalized evaluation of metasurface structures. We use freeform dielectric metasurfaces as a model system and demonstrate with MetaChat the design of multi-objective, multi-wavelength metasurfaces orders of magnitude faster than conventional methods. These concepts present a scientific computing blueprint for utilizing specialist design agents, surrogate solvers, and human interactions to drive multi-physics innovation and discovery.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 34 of its 107 references that are in the map. First posted on arXiv 26 Mar 2025.

Builds on

34 of 107 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2022Training language models to follow instructions with human feedbackcited by 15,971
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  6. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  7. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  8. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  9. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  10. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  11. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,908
  12. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
22 more
  1. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  2. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  3. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,184
  4. 2023A Survey on Multimodal Large Language Modelscited by 909
  5. 2023LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Datasetcited by 461
  6. 2022Fourier Neural Operator with Learned Deformations for PDEs on General Geometriescited by 427
  7. 2025Accelerating scientific discovery with Co-Scientistcited by 401
  8. 2021U-FNO -- An enhanced Fourier neural operator-based deep-learning model for multiphase flowcited by 297
  9. 2021Physics-informed neural networks with hard constraints for inverse designcited by 280
  10. 2020hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decompositioncited by 236
  11. 2018All-Optical Machine Learning Using Diffractive Deep Neural Networkscited by 177
  12. 2019PPINN: Parareal Physics-Informed Neural Network for time-dependent PDEscited by 118
  13. 2020Deep neural networks for the evaluation and design of photonic devicescited by 50
  14. 2024AtomAgents: Alloy design and discovery through physics-aware multi-modal multi-agent artificial intelligencecited by 36
  15. 2019Global optimization of dielectric metasurfaces using a physics-driven neural networkcited by 30
  16. 2023OptoGPT: A Foundation Model for Inverse Design in Optical Multilayer Thin Film Structurescited by 13
  17. 2024PreAct: Prediction Enhances Agent's Planning Abilitycited by 12
  18. 2024Towards General Neural Surrogate Solvers with Specialized Neural Acceleratorscited by 9
  19. 2024Learning Electromagnetic Metamaterial Physics With ChatGPTcited by 6
  20. 2022DeepAdjoint: An All-in-One Photonic Inverse Design Framework Integrating Data-Driven Machine Learning with Optimization Algorithmscited by 5
  21. 2022End-to-End Optimization of Metasurfaces for Imaging with Compressed Sensingcited by 4
  22. 2020Heuristic Methods and Performance Bounds for Photonic Designcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 7.

14 papers.

Sort:

By area:

  1. 2025An Agentic Framework for Autonomous Metamaterial Modeling and Inverse Designcited by 8
  2. 2025Inverse Design in Nanophotonics via Representation Learningcited by 7
  3. 2025ATHENA: Agentic Team for Hierarchical Evolutionary Numerical Algorithmscited by 7
  4. 2026GRACE: an Agentic AI for Particle Physics Experiment Design and Simulationcited by 6
  5. 2025Accurate and scalable deep Maxwell solvers using multilevel iterative methodscited by 5
  6. 2025Chat to Chip: Large Language Model Based Design of Arbitrarily Shaped Metasurfacescited by 4
  7. 2026A Self-Evolving Agentic Framework for Metasurface Inverse Designcited by 3
  8. 2025SelfAI: A self-directed framework for long-horizon scientific discoverycited by 2
  9. 2025Research and Prototyping Study of an LLM-Based Chatbot for Electromagnetic Simulationscited by 0
  10. 2026A Differentiable Ray-Wave Framework for Hybrid Refractive-Diffractive System Modeling and Optimizationcited by 0
  11. 2026Agentic Root Cause Analysis through Evidence-Grounded Reasoningcited by 0
  12. 2026VortexChat: An agentic framework for autonomous multi-objective integrated photonic designcited by 0
2 more
  1. 2026Towards a universal meta-optics solver via large language modelscited by 0
  2. 2026HALO: A Physics-Aware LLM Agent Framework for Nanophotonic Designcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026A Comprehensive Review of Large Language Models for Nanophotonics: From Surrogate Modeling to Autonomous Design
  2. 2025Meta-GPT: Decoding the Metasurface Genome with Generative Artificial Intelligence
  3. 2025A Multi-Agent Framework Integrating Large Language Models and Generative AI for Accelerated Metamaterial Design
  4. 2025AI Agents for Photonic Integrated Circuit Design Automation
  5. 2026METASYMBO: Multi-Agent Language-Guided Metamaterial Discovery via Symbolic Latent Evolution
  6. 2026OPTIAGENT: A Physics-Driven Agentic Framework for Automated Optical Design
  7. 2024Self-driving lab discovers principles for steering spontaneous emission
  8. 2025Multi-Agent Reinforcement Learning for Inverse Design in Photonic Integrated Circuits