PaperSeal

Inspiration

We wanted to make it easier for students and researchers to judge whether a source is actually trustworthy. Finding sources is easy, but evaluating credibility still takes time and experience.

What it does

PaperSeal is a Chrome extension that analyzes the current research webpage and gives a credibility verdict: Credible, Questionable, or Weak. It also provides a score breakdown, explanation, and any red flags it finds.

How we built it

We built PaperSeal as a Chrome extension that extracts visible page content from the DOM, pulls metadata such as citations and venue from Semantic Scholar, uses an LLM only for signal extraction, and then applies a deterministic scoring system for the final verdict.

Challenges we ran into

One challenge was keeping the system explainable and consistent by separating LLM analysis from final scoring. Another was dealing with inconsistent webpage structure and extracting useful research signals from noisy page content.

Accomplishments that we're proud of

One challenge was keeping the system explainable and consistent by separating LLM analysis from final scoring. Another was dealing with inconsistent webpage structure and extracting useful research signals from noisy page content.

What we learned

We learned that credibility assessment works best when the system is transparent, constrained, and easy to explain. We also learned how important it is to clearly separate extraction, metadata enrichment, and scoring.

What's next for PaperSeal

Next, we want to improve extraction reliability, support a wider range of research webpages, refine the scoring rubric, and make the explanations even more useful for students doing literature review.

Built With

• chrome-extension-api
• dom-extraction
• groq
• html/css
• javascript
• llm
• semantic-scholar-api