Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Virtual Scientific Companion for Synchrotron Beamlines: A Prototype (2312.17180v1)

Published 28 Dec 2023 in cs.CL and cs.AI

Abstract: The extraordinarily high X-ray flux and specialized instrumentation at synchrotron beamlines have enabled versatile in-situ and high throughput studies that are impossible elsewhere. Dexterous and efficient control of experiments are thus crucial for efficient beamline operation. Artificial intelligence and machine learning methods are constantly being developed to enhance facility performance, but the full potential of these developments can only be reached with efficient human-computer-interaction. Natural language is the most intuitive and efficient way for humans to communicate. However, the low credibility and reproducibility of existing LLMs and tools demand extensive development to be made for robust and reliable performance for scientific purposes. In this work, we introduce the prototype of virtual scientific companion (VISION) and demonstrate that it is possible to control basic beamline operations through natural language with open-source LLM and the limited computational resources at beamline. The human-AI nature of VISION leverages existing automation systems and data framework at synchrotron beamlines.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (47)
  1. High-resolution non-destructive three-dimensional imaging of integrated circuits. Nature, 543(7645):402–406, 2017.
  2. Three-dimensional imaging of integrated circuits with macro-to nanoscale zoom. Nature Electronics, 2(10):464–470, 2019.
  3. Correlated x-ray 3d ptychography and diffraction microscopy visualize links between morphology and crystal structure of lithium-rich cathode materials. IScience, 11:356–365, 2019.
  4. Three-dimensional imaging of biological tissue by cryo x-ray ptychography. Scientific reports, 7(1):1–12, 2017.
  5. Alterations in sub-axonal architecture between normal aging and parkinson’s diseased human brains using label-free cryogenic x-ray nanotomography. Frontiers in neuroscience, page 1152, 2020.
  6. Light-activated contraction in organic-inorganic 2d perovskites enables high-efficiency photovoltaics. In Materials Research Society-Spring Meeting 2021 (MRS 2021 Spring Meeting), 2021.
  7. Detlef-M Smilgies. GISAXS: A versatile tool to assess structure and self-assembly kinetics in block copolymer thin films. Journal of Polymer Science, 60(7):1023–1041, 2022.
  8. Large-grained cylindrical block copolymer morphologies by one-step room-temperature casting. Macromolecules, 53(24):11178–11189, 2020.
  9. Solvent-free coating of organic semiconductor membranes with centimetric crystalline domains. Advanced Electronic Materials, 7(3):2000792, 2021.
  10. Crystal structure and orientation of organic semiconductor thin films by microcrystal electron diffraction and grazing-incidence wide-angle x-ray scattering. Chemical Communications, 56(30):4204–4207, 2020.
  11. Esther Tsai. Virtual scientific companion for synchrotron beamlines, 2023. https://science.osti.gov/-/media/early-career/pdf/FY-2023-DOE-SC-Early-Career-Research-Program-Abstracts.pdf [Accessed: 2023-12].
  12. Autonomous chemical research with large language models. Nature, 624(7992):570–578, 2023.
  13. Leveraging large language models for predictive chemistry. 2023.
  14. Assessment of chemistry knowledge in large language models that generate code. Digital Discovery, 2(2):368–376, 2023.
  15. Kevin G Yager. Domain-specific chatbots for science using embeddings. Digital Discovery, 2(6):1850–1861, 2023.
  16. Opportunities for retrieval and tool augmented large language models in scientific facilities. arXiv preprint arXiv:2312.01291, 2023.
  17. Adaptive 3d convolutional neural network-based reconstruction method for 3d coherent diffraction imaging. Journal of Applied Physics, 128(18):184901, 2020.
  18. wav2vec 2.0: A framework for self-supervised learning of speech representations. Advances in neural information processing systems, 33:12449–12460, 2020.
  19. Robust speech recognition via large-scale weak supervision. arXiv preprint arXiv:2212.04356, 2022.
  20. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901, 2020.
  21. Improving language understanding by generative pre-training. 2018.
  22. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019.
  23. Gpt-4 technical report, 2023.
  24. BERT: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805, 2018.
  25. LaMDA: Language models for dialog applications, 2022.
  26. Palm: Scaling language modeling with pathways, 2022.
  27. Training compute-optimal large language models. arXiv preprint arXiv:2203.15556, 2022.
  28. Galactica: A large language model for science. arXiv preprint arXiv:2211.09085, 2022.
  29. Llama: Open and efficient foundation language models, 2023.
  30. Mistral 7B, 2023.
  31. Google’s neural machine translation system: Bridging the gap between human and machine translation. arXiv preprint arXiv:1609.08144, 2016.
  32. Attention is all you need. Advances in neural information processing systems, 30, 2017.
  33. BERT base model (uncased) on HuggingFace. https://huggingface.co/bert-base-uncased [Accessed: 2023-12].
  34. Huggingface’s transformers: State-of-the-art natural language processing, 2020.
  35. Wilson L Taylor. “cloze procedure”: A new tool for measuring readability. Journalism quarterly, 30(4):415–433, 1953.
  36. Bluesky’s Ahead: A Multi-Facility Collaboration for an a la Carte Software Project for Data Acquisition and Management. Synchrotron Radiation News, 32(3):19–22, 2019.
  37. VISION prototype demo video at NSLS-II CMS, 2023. https://drive.google.com/file/d/1mqEY43Uik2zYH-YAmixR8OHy-6JnwOpJ/view?usp=drive_link [Accessed: 2023-12].
  38. ReAct: Synergizing reasoning and acting in language models, 2023.
  39. Toolformer: Language models can teach themselves to use tools. arXiv preprint arXiv:2302.04761, 2023.
  40. PAL: Program-aided language models, 2023.
  41. GPTQ: Accurate post-training quantization for generative pre-trained transformers, 2023.
  42. Georgi Gerganov. Ggml, 2023. https://github.com/ggerganov/ggml [Accessed: 2023-12].
  43. The case for 4-bit precision: k-bit inference scaling laws, 2023.
  44. Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning. Advances in Neural Information Processing Systems, 35:1950–1965, 2022.
  45. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685, 2021.
  46. Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities. Nature Reviews Physics, 3(10):685–697, 2021.
  47. Autonomous discovery of emergent morphologies in directed self-assembly of block copolymer blends. Science Advances, 9(2):eadd3687, 2023.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Youtube Logo Streamline Icon: https://streamlinehq.com

YouTube