I built RenderScholar, a small tool for researchers and students who are tired of LLMs hallucinating citations.
RenderScholar scrapes Google Scholar directly, pulls real papers, and then ranks them by relevance, citations, and recency. It also renders results into two views:
- Human view: a clean HTML page for browsing papers
- LLM view: copy-paste friendly citations that you can safely feed into a language model
Why I made it: I kept getting fake references from ChatGPT/Claude when doing literature reviews. I wanted a quick way to gather real citations that I could trust, and then drop them into my workflow.
I’d love feedback—especially from people who would use this for lit reviews, etc and are frustrated with hallucinated citations. Also curious if there are features that would make it more useful.
peterdunson•1h ago
RenderScholar scrapes Google Scholar directly, pulls real papers, and then ranks them by relevance, citations, and recency. It also renders results into two views:
- Human view: a clean HTML page for browsing papers - LLM view: copy-paste friendly citations that you can safely feed into a language model
Why I made it: I kept getting fake references from ChatGPT/Claude when doing literature reviews. I wanted a quick way to gather real citations that I could trust, and then drop them into my workflow.
Repo: https://github.com/peterdunson/renderscholar
I’d love feedback—especially from people who would use this for lit reviews, etc and are frustrated with hallucinated citations. Also curious if there are features that would make it more useful.