News Summarization with Open Source Large Language Models
During the Spring 2024, we completed a project with Rystad Energy focusing on harnessing the power of both private and open-source Large Language Models (LLMs) to automate time-consuming tasks.
Our exploration included cutting-edge models like Llama-2, Llama-3, Gemma and GPT-4, aiming to enhance text summarization of news articles, a task previously done by humans. By constructing specialized evaluation metrics, we estimated and compared the quality of different models, tailored to Rystad Energy’s use case.
The results? Our fine-tuned models produced concise summaries containing relevant entities and excelled in pinpointing energy-sector specifics from extensive texts. One of our models is also currently being used in production by Rystad Energy.
Our delivery consisted of the code used to produce and test new models, as well as a report on our findings, containing details on performance, quality, and economic viability.
-
Rystad Energy is a leading global energy research and business intelligence firm, renowned for its vast databases and in-depth analysis across the oil, gas, and renewable energy sectors. Since its beginning 20 years ago, Rystad has delivered consulting and analytics services to a wide array of entities and is present all over the world, with offices in Oslo, New York, London, Singapore, Rio De Janeiro, Beijing and Sydney.