AI Science Tracker
EN

National Institute of Advanced Industrial Science and Technology

1 discovery from 2026

Discoveries

Breaking the Fixed-Substrate Limitation of Deep Learning Ellipsometry with Pseudo-Substrate Spectra

Source check: Source checked. Not independent scientific confirmation.

Finding / tested result: FiLM-DLE takes both a thin-film spectrum and a pre-deposition pseudo-substrate spectrum as inputs. On 47 measured films across diverse substrates, its optical-constant and thickness estimates had mean absolute errors of 0.047, 0.030 and 1.52 nm, respectively, against manual analysis; each analysis took under two seconds.

Technology: Two-input deep-learning ellipsometry using pseudo-substrate spectra

Publication: Preprint · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
FiLM-DLE is a neural network that takes a thin-film spectrum and a measured pre-deposition substrate spectrum as inputs to estimate film optical constants and thickness.
Limitation / still unknown
The study evaluates the reported models and datasets only; broader generalization and real-world deployment outcomes were not established.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/pdf/2611.20694

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: linked separately.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.