A Virtual Assistant for Cybersickness Care
Bibliographic record
Abstract
We present an avatar and task-oriented dialog agent for monitoring user discomfort during a virtual reality (VR) cognitive exercise and providing personalized information and advice on its relief. The goal of this approach is to provide instantaneous assistance to users for a more comfortable VR experience, thereby enabling them to spend more time on cognitive tasks. We developed an avatar in a VR environment with which users may communicate verbally, and a dialog agent in a machine-learning based conversational AI platform. We performed a technical evaluation of the natural language understanding (NLU) component by comparing 2 models (BERT and StarSpace) using a train-test split, showing a significant benefit of BERT with smaller data sets. We validated the turn prediction using a train-test split and using randomly generated conversations. Both validations showed acceptable conversation-level accuracy. We undertook a usability study at two sites, showing effectiveness at both and good acceptability at one of the two. The framework outlined can be used to develop other virtual agents for cognitive self-care. Suggested improvements include validating the avatar with integrated BERT and reducing reliance on data augmentation, offline voice interaction modules, improved UX design, clinically validating the effect of the dialog agent on user discomfort and on cognitive performance, and increasing the ubiquity of the avatar within the VR cognitive care environment.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.015 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".