A5052773688- Oral and Maxillofacial Pathology
- Oral Health Pathology and Treatment
- Topic Modeling
- Autonomous Vehicle Technology and Safety
- Sinusitis and nasal conditions
- Dysphagia Assessment and Management
- Head and Neck Surgical Oncology
- Gastroesophageal reflux and treatments
- Transportation Safety and Impact Analysis
- Cervical Cancer and HPV Research
- Tumors and Oncological Cases
- Respiratory and Cough-Related Research
- Head and Neck Cancer Studies
- Web Application Security Vulnerabilities
- Adversarial Robustness in Machine Learning
- Vehicular Ad Hoc Networks (VANETs)
Era's Lucknow Medical College and Hospital
2018
As Large Language Models (LLMs) and generative AI become increasingly widespread, concerns about content safety have grown in parallel. Currently, there is a clear lack of high-quality, human-annotated datasets that address the full spectrum LLM-related risks are usable for commercial applications. To bridge this gap, we propose comprehensive adaptable taxonomy categorizing risks, structured into 12 top-level hazard categories with an extension to 9 fine-grained subcategories. This designed...
Controversy surrounds regarding the role of human papillomavirus (HPV) in oral precancerous and cancerous lesions India where smokeless, tobacco consumption is rampant.The present study was carried out with an aim to investigate presence type HPV infection oropharyngeal leukoplakia determine association positivity various patient lesion characteristics.Prospective case series.A total fifty confirmed cases (OL), aged 18-55 years were enrolled study. Specimens obtained from site lesion. All...
As Large Language Models (LLMs) and generative AI become more widespread, the content safety risks associated with their use also increase. We find a notable deficiency in high-quality datasets benchmarks that comprehensively cover wide range of critical areas. To address this, we define broad risk taxonomy, comprising 13 9 sparse categories. Additionally, curate AEGISSAFETYDATASET, new dataset approximately 26, 000 human-LLM interaction instances, complete human annotations adhering to...
<p class="abstract"><strong>Background:</strong> The objective of the study was to spectrum premalignant oral lesions in rural North Indian population at a tertiary care hospital.</p><p class="abstract"><strong>Methods:</strong> A total 340 patients with for more than 2 weeks were enrolled study. Patients proven malignancy, white patch cavity which can be scrapped off, e.g. candidiasis and urban excluded. Biopsy specimen obtained from all evaluated...