05 / Reference

Glossary and references

Terminology and sources used throughout the report.

BIC

A mode embedded in the radiation continuum that nevertheless does not radiate because symmetry or interference cancels leakage.

Quasi-BIC

A practical, slightly leaky version of a BIC: still high-Q, but now experimentally excitable and readable.

Reservoir computing

A computational framework where a complex dynamical system transforms inputs into a rich state space and only a simple readout is trained.

Supermode

A collective eigenmode of the coupled system rather than of a single isolated node.

Gain saturation

The nonlinear response that appears as stimulated emission depletes the available gain, producing a thresholded transfer change.

PPF

Paired-pulse facilitation. The second pulse can evoke a stronger response because the first pulse leaves useful residual excitation behind.

Injection locking

A way to synchronize or control lasers with other optical signals, useful for cascaded photonic systems.

SLM / DMD

Spatial light modulator or digital micromirror device. Practical tools for projecting many independently programmed pump patterns onto a photonic substrate.

02 / References

Primary papers and official sources

Selected sources used throughout the report, weighted toward primary papers and official company material.

Fu, J., Jin, R., Xie, Z., Tang, H. et al. Photonic Neuromorphic Computing enabled by a BIC Metasurface. arXiv:2602.22528 (2026). arXiv
Tang, H., Huang, C., Wang, Y. et al. Dynamically tunable long-range coupling enabled by bound state in the continuum. Light: Science & Applications 14, 278 (2025). DOI
Kang, M., Liu, T., Chan, C. T., Xiao, M. Applications of Bound States in the Continuum in Photonics. Nature Reviews Physics 5, 659-678 (2023). DOI
Yan, M., Huang, C., Bienstman, P. et al. Emerging opportunities and challenges for the future of reservoir computing. Nature Communications 15, 2056 (2024). DOI
Lukosevicius, M., Jaeger, H. Reservoir computing approaches to recurrent neural network training. Computer Science Review 3(3), 127-149 (2009). DOI
Tanaka, G., Yamane, T., Heroux, J. B. et al. Recent advances in physical reservoir computing: A review. Neural Networks 115, 100-123 (2019). DOI
Shastri, B. J., Tait, A. N., Ferreira de Lima, T. et al. Photonics for artificial intelligence and neuromorphic computing. Nature Photonics 15, 102-114 (2021). DOI
Chen, Z., Sludds, A., Davis III, R. et al. Deep learning with coherent VCSEL neural networks. Nature Photonics 17, 723-730 (2023). DOI
Shen, Y. W., Li, R. Q., Liu, G. T. et al. Deep photonic reservoir computing recurrent network. Optica 10(12), 1745-1751 (2023). arXiv
Ji, K., Tirabassi, G., Masoller, C. et al. Photonic neuromorphic computing using symmetry-protected zero modes in coupled nanolaser arrays. Nature Communications 16, 9203 (2025). DOI
Gan, Y., Shi, Y., Ghosh, S. et al. Ultrafast neuromorphic computing driven by polariton nonlinearities. eLight 5, 9 (2025). DOI
Shahroudy, A. et al. NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis. CVPR (2016). arXiv
Chakrabarty, N. Brain MRI Images for Brain Tumor Detection. Kaggle dataset (2018). Kaggle
Intel Silicon Photonics. Official product page. Includes Intel's claim of more than 8 million PICs and over 32 million on-chip lasers shipped for data-center networking. Intel
Optical I/O Solutions Optimized for AI Workloads. Ayar Labs official page positioning optical I/O as essential for AI hardware efficiency. Ayar Labs
Lightmatter Unveils Passage M1000 Photonic Superchip, World's Fastest AI Interconnect. Lightmatter press release on a photonic interposer for next-generation AI infrastructure. Lightmatter
Broadcom Announces Third-Generation Co-Packaged Optics Technology with 200G/lane Capability. Broadcom press release on CPO for AI networking. Broadcom
NVIDIA Announces Spectrum-X Photonics, Co-Packaged Optics Networking Switches to Scale AI Factories to Millions of GPUs. NVIDIA press release on silicon-photonics networking switches for AI factories. NVIDIA
Marvell Completes Acquisition of Celestial AI. Marvell press release on bringing Celestial AI's Photonic Fabric into Marvell's AI-infrastructure portfolio. Marvell
AMD Acquires Enosemi to Accelerate Co-Packaged Optics Innovation for AI Systems. AMD blog post on the Enosemi acquisition and photonics roadmap. AMD
iPronics About Us / PROMETHEUS. Company page noting PROMETHEUS and the development of photonic spiking neural networks and neuromorphic architectures. iPronics
PACE 2. Lightelligence product page for its optical computing platform. Lightelligence
Akhetonics. Company page describing an all-optical XPU for general-purpose compute and AI. Akhetonics
NVIDIA Announces Strategic Partnership With Lumentum to Develop State-of-the-Art Optics Technology. Official announcement of a multiyear optics partnership and $2B NVIDIA investment in Lumentum. Lumentum
NVIDIA and Coherent Announce Strategic Partnership to Develop Optics Technology to Scale Next-Generation Data Center Architecture. Official announcement of a multiyear partnership between NVIDIA and Coherent. Coherent
What We Do. Optalysys overview page on optical computing as a commercial adjacency relevant to photonic compute architectures. Optalysys

03 / Source notes

How the source set is used

The report combines different source types, but keeps them separated conceptually.

Primary paper

Main technical claim

The report treats Fu et al. as the source of record for the device, experiment, and benchmark results.

Adjacent literature

Interpretive context

Reservoir-computing reviews and neighboring photonic systems are used to interpret what the result implies and what it does not.

Official company pages

Infrastructure layer

Company and investor pages are used only for present-tense placement of the broader photonic stack.

Selection rule

Preference for traceability

Where possible, the report prefers sources that make it easy to verify claims directly rather than rely on secondary summaries.