AI Intern

The candidate will work in high visibility projects as a Data Scientist, bringing the Data Science and NLP expertise to projects. The candidate will work in RD Data Science team and collaborate with Product’s managers, domain experts, Knowledge representation experts, to build high value outcome from Elsevier content. The candidate will have an opportunity to impact virtually all Elsevier applications related to Research and Operations.

Primary Objectives: Developing a Multi-Agent LLM Framework: Establish a system where

multiple LLM agents collaborate to generate high-quality topic pages.

Outline Generation LLM: Create an agent responsible for generating a coherent and

comprehensive outline for each topic page. Creating an outline mimics the pre-writing stage in

human writing process.

Research and Summary Writing: Implement a RAG setting where an agent retrieves relevant

scientific snippets and writes detailed summaries for each section of the outline. This involves

querying scientific databases, extracting key information, and synthesizing it into coherent

sections. In addition to the summary, this agent will also retrieve metadata such as prolific authors,

journals, and trending topics around each concept and write a summary of them as well.

Quality Evaluation and Feedback: Design an agent to evaluate the generated content (accuracy,

completeness, coherence, and readability) and provide constructive feedback for refinement.

Evaluation Criteria:

Accuracy: The correctness and reliability of the information presented in the generated topic pages.

This will be measured by means of human and automatic evaluation. The automatic evaluation

will focus on the factuality of the content with respect to the snippets.

Comprehensiveness: The extent to which the topic pages cover all relevant aspects of the scientific

concepts. This will be measured by means of human and automatic evaluation. The automatic

evaluation will focus on the coverage of the generated content to the topics discussed in all snippets

available per concept.

Readability: The clarity and coherence of the content. This aspect will also be evaluated by

means of human and automatic evaluation. The automatic evaluation will be done using an LLM.

User Satisfaction: Feedback from users regarding the utility and quality of the generated topic

pages. This aspect will be evaluated by subject matter experts.

———————————————————————–

Elsevier is an equal opportunity employer: qualified applicants are considered for and treated during employment without regard to race, color, creed, religion, sex, national origin, citizenship status, disability status, protected veteran status, age, marital status, sexual orientation, gender identity, genetic information, or any other characteristic protected by law. We are committed to providing a fair and accessible hiring process. If you have a disability or other need that requires accommodation or adjustment, please let us know by completing our Applicant Request Support Form: https://forms.office.com/r/eVgFxjLmAK , or please contact 1-855-833-5120.

Please read our Candidate Privacy Policy.