# Released website examples

Eight text records copied from the public PsycheMME review subset, selected for explicit observable cues, temporal/history links and alignment with the capability definitions. Question/reference contents and review flags are preserved; no media is copied. These are redacted public copies, not the original frozen evaluation inputs.

# Source attribution and permissions

These example assets redistribute text only and add no video, audio file, screenshot or participant photograph. Third-party licenses govern the available text-derived data; source locators do not grant new rights.

## EasyCom

EasyCom: An Augmented Reality Dataset to Support Algorithms for Easy Communication in Noisy Environments, Jacob Donley, Vladimir Tourbabin, Jung-Suk Lee, Mark Broyles, Hao Jiang, Jie Shen, Maja Pantic, Vamsi Krishna Ithapu and Ravish Mehra (2021). Original source and licensing: https://github.com/facebookresearch/EasyComDataset and https://github.com/facebookresearch/EasyComDataset/blob/main/LICENSE .

The 70 EasyCom-derived questions, associated references and 40 episode transcripts in `dataset/text_subset/` are distributed under CC BY-NC 4.0. Changes include question/annotation creation, transcription, ID pseudonymization and text redaction. Noncommercial use and original attribution remain required. No endorsement is implied.


## Website selection update (2026-10-17)

Selected IDs: Q0348, Q0351, Q0876, Q0368, Q0453, Q0749, Q0609, Q0829. No model performance was used. Compared with the earlier file-order preview, G now illustrates a shared activity, A a physical orientation/space-making policy, M an explicit HISTORY/CURRENT link, S a quoted clarification, and R a quoted warning. Exact public inputs/references and their review metadata are preserved; neither benchmark answers nor paper results were changed. This is a text-based presentation review, not a new video-level annotation audit.