- Built an LLM-powered application that can summarize any research paper into a concise summary, extract tables from the research paper, and give a real-time evaluation score for the generated summary.
- Utilized groq’s inference for Meta’s LLaMA 3–8B to generate summaries, evaluated them using BERTScore (F1) and ROUGE, and extracted tables with pdfplumber and pandas.
- Achieved an average BERTScore (F1) of 0.87 across a benchmark of 10+ scientific papers, reflecting strong semantic alignment with original content.
-
Notifications
You must be signed in to change notification settings - Fork 0
An LLM-powered application that summarizes scientific research papers, extracts tables, and provides real-time evaluation using BERTScore (F1) and ROUGE. Built using Meta’s LLaMA 3–8B via Groq, with table extraction powered by pdfplumber and pandas.
TSS-sniper/Research-paper-Summarizer-with-Realtime-Eval
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
An LLM-powered application that summarizes scientific research papers, extracts tables, and provides real-time evaluation using BERTScore (F1) and ROUGE. Built using Meta’s LLaMA 3–8B via Groq, with table extraction powered by pdfplumber and pandas.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published