Papers by Prajvi Saxena
Newspaper Signaling for Crisis Prediction (2024.naacl-demo)
Copied to clipboard
| Challenge: | Existing systems for detecting crisis-related signals are limited due to unstructured data, media, and cultural bias, and multiple languages. |
| Approach: | They propose a model for multi-lingual and open-domain newspaper signaling for detecting crisis-related indicators in newspaper articles. |
| Outcome: | The proposed model can detect crisis-related indicators in multiple languages and can be used in open crisis domains in real-time. |
Streamlining LLMs: Adaptive Knowledge Distillation for Tailored Language Models (2025.naacl-srw)
Copied to clipboard
| Challenge: | Large language models (LLMs) have transformative potential across industries, e.g., enhancing customer service, revolutionizing medical diagnostics, or identifying crises in news articles. |
| Approach: | They propose to distill compact, parameter-efficient tailored language models from LLMs for domain-specific tasks with comparable performance. |
| Outcome: | The proposed framework outperforms knowledge distillation frameworks in the crisis domain, where labeled data is scarce. |