arXiv subjects

Jordan Lee

Publications and source records attributed to Jordan Lee.

3 recordsLinked to original sources

arXiv2026

AutoSpecNER: A Fine-Grained Named Entity Recognition Dataset for Vehicle Specification Extraction

Vehicle advertisements contain rich specification information, but automotive NER resources remain limited. We introduce AutoSpecNER, an expert-annotated dataset for fine-grained entity recognition in vehicle listings. The dataset includes 659 advertisements from a popular car-selling website, with over 10,000 entities annotated across 15 categories, including MODEL, ENGINE_SPEC, and BATTERY_CAPACITY. Annotation quality was validated through inter-annotator agreement, achieving an average score of 91.5%. We benchmark rule-based extraction, fine-tuned transformer encoders, and large language models. DeBERTa achieves the best performance with a 90% micro-F1 score, outperforming the rule-based baseline (43%) and the strongest large language model (77.8%).

cs.CL↗

arXiv2023

Gravitational waves from high-power twisted light

Recent advances in high-energy and high-peak-power laser systems have opened up new possibilities for fundamental physics research. In this work, the potential of twisted light for the generation of gravitational waves in the high frequency regime is explored for the first time. Focusing on Bessel beams, novel analytic expressions and numerical computations for the generated metric perturbations and associated powers are presented. Compelling evidence is provided that the properties of the generated gravitational waves, such as frequency, polarisation states and direction of emission, are controllable by the laser pulse parameters and optical arrangements.

gr-qc↗

arXiv2021

Identifying Human Edited Images using a CNN

Most non-professional photo manipulations are not made using propriety software like Adobe Photoshop, which is expensive and complicated to use for the average consumer selfie-taker or meme-maker. Instead, these individuals opt for user friendly mobile applications like FaceTune and Pixlr to make human face edits and alterations. Unfortunately, there is no existing dataset to train a model to classify these type of manipulations. In this paper, we present a generative model that approximates the distribution of human face edits and a method for detecting Facetune and Pixlr manipulations to human faces.

cs.CV↗