Shasta Ihorn

Shasta Ihorn

Auto-generated profile, based on publicly available information

Shasta Ihorn

Department of Psychology, San Francisco State University, United States

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 October 2023
Verified
Conference Paper
Open access
ASAbigale Stangl
Abigale Stangl
SIShasta Ihorn
Shasta Ihorn
YSYue-Ting Siu
Yue-Ting Siu
ABAditya Bodi
Aditya Bodi
MCMar Castanon
Mar Castanon
IYIlmi Yoon
Ilmi Yoon

26 January 2018 in Journal of Educational and Psychological Consultation
Verified
Journal Article
Closed access
SIShasta M. Ihorn
Shasta M. Ihorn
PAPrerna Arora
Prerna Arora
Teleconsultation is being increasingly used in schools to support children with disabilities; however, further evaluation of the impact of this practice is needed. This study describes a teleconsultation model used to provide educational support for students with visual impairments being educated in their local school districts and examines the initial acceptability and feasibility of this teleconsultation model. In addition, the current study discusses some of the recommendations for improvement regarding program implementation, with the goal of increasing the ease of use of teleconsultation in schools. Stakeholders reported qualitative and quantitative data indicating that the teleconsultation model is acceptable to and feasible for families and school district staff.

19 April 2023
Verified
Conference Paper
Open access
ASAndrew Taylor Scott
Andrew Taylor Scott
LNLothar D Narins
Lothar D Narins
AKAnagha Kulkarni
Anagha Kulkarni
MCMar Castanon
Mar Castanon
BKBenjamin Kao
Benjamin Kao
IYIlmi Yoon
Ilmi Yoon
How well a caption fits an image can be difficult to assess due to the subjective nature of caption quality. What is a good caption? We investigate this problem by focusing on image-caption ratings and by generating high quality datasets from human feedback with gamification. We validate the datasets by showing a higher level of inter-rater agreement, and by using them to train custom machine learning models to predict new ratings. Our approach outperforms previous metrics – the resulting datasets are more easily learned and are of higher quality than other currently available datasets for image-caption rating.

5 March 2026
Verified
Conference Paper
Open access
JBJuvenal Francisco Barajas
Juvenal Francisco Barajas
GJGio Jung
Gio Jung
MSManali Seth
Manali Seth
LDLana Do
Lana Do
MJMalleeswari Jagabattuni
Malleeswari Jagabattuni
IYIlmi Yoon
Ilmi Yoon
Large language models (LLMs) can generate synthetic video captions at scale, offering a practical way to enrich datasets that lack multiple human references. However, evaluating the semantic reliability of such captions is challenging, particularly when traditional lexical-overlap metrics (e.g., BLEU, ROUGE, METEOR, CIDEr, SPICE) are sensitive to surface-level wording and often misalign with human judgments. This paper investigates whether sentence embedding models provide more reliable evaluation metrics for LLM-generated video captions. We first compare five embedding models—Sentence-BERT, SimCSE, MPNet, and two GPT-based embeddings—against traditional lexical metrics using 450 caption pairs from VATEX, a benchmark video captioning dataset. Through validation with a linguistic specialist, we find that sentence embedding-based approaches achieve substantially higher agreement with expert judgments using Kendall’s τ correlation. We then used GPT-4o to generate ten synthetic captions per video from VATEX and apply the validated embedding metrics to compare semantic similarities with the human-human caption pairs. We applied the same pipeline to YuWa, a single-reference accessibility video dataset, which demonstrates a real-world use case. Our findings suggest that synthetic captions achieve more consistency in semantic similarity levels compared to independently written human captions. Notably, augmenting YuWa’s single human caption per video with ten LLM-generated captions produces semantic similarity scores matching or surpassing those of VATEX’s ten independently written human captions. This work provides the first expert-validated framework for evaluating LLM-generated video captions and demonstrates a practical pipeline for quality-controlled dataset augmentation in accessibility research.

1 January 2009
Verified
Dataset
Closed access
PAPrerna Arora
Prerna Arora
DTDeborah J. Tharinger
Deborah J. Tharinger
LJLaura E. Judd
Laura E. Judd
JWJudith Wan
Judith Wan
SIShasta M. Ihorn
Shasta M. Ihorn

24 October 2026
Verified
Conference Paper
Open access
LDLana Do
Lana Do
GJGio Jung
Gio Jung
ASAndrew Taylor Scott
Andrew Taylor Scott
JBJuvenal Francisco Barajas
Juvenal Francisco Barajas
ABAlexander Mario Blum
Alexander Mario Blum
IYIlmi Yoon
Ilmi Yoon