ASAndrew Taylor Scott

Andrew Taylor Scott

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Andrew Taylor Scott has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Toward Scalable Audio Description Quality Control: A Workflow for Evaluating Human and VLM Raters

    by Lana Do, Gio Jung, Andrew Taylor Scott, Juvenal Francisco Barajas, Alexander Mario Blum, Shasta Ihorn, Vassilis Athitsos, Ilmi Yoon

  • Open Access

    Beyond Lexical Overlap: Validating Semantic Similarity Metrics for LLM-Based Video Caption Augmentation

    by Juvenal Francisco Barajas, Gio Jung, Manali Seth, Lana Do, Malleeswari Jagabattuni, Andrew Taylor Scott, Shasta Ihorn, Alexander Mario Blum, Vassilis Athitsos, 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.

  • Open Access

    Improved Image Caption Rating – Datasets, Game, and Model

    by Andrew Taylor Scott, Lothar D Narins, Anagha Kulkarni, Mar Castanon, Benjamin Kao, Shasta Ihorn, Yue-Ting Siu, 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.